Cleaner vocal synth: less static, click-free note transitions

- Jitter reduced (0.3% → 0.1%), shimmer reduced (2% → 0.8%)
- Breath noise halved (0.08 → 0.04), mix 85/15 → 92/8
- 10ms fade in/out on every vocal note prevents clicks
- Smoother syllable transitions

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

CHANGELOG.md

@@ -2,6 +2,30 @@
 
 All notable changes to PyTheory are documented here.
 
+## 0.35.0
+
+- **8.5x faster import** — dropped pytuning/sympy, lazy-load scipy.
+  `import pytheory` now takes ~50ms instead of ~480ms (#44)
+- **Proper shruti JI ratios** — 22 positions with 5-limit just intonation
+  (pure 3/2 fifths, 5/4 thirds), not 22-TET approximation
+- **Arabic maqam JI ratios** — Zalzalian 11-limit ratios.
+  Mi↓ (the Rast third) is exactly 27/22 from Do
+- **B#/Cb octave boundary fix** — B#4 = C5, Cb4 = B3 (#45)
+- **Int tone names** — `Tone(0, system=TET(22))` works alongside strings.
+  Wrapping: `Tone(22)` → tone 0, octave+1. `System.tone()` convenience.
+- **Timpani synth** — inharmonic membrane modes, felt mallet, copper kettle
+  resonance, cathedral reverb
+- **Saxophone synth** — conical bore, reed buzz, brass body warmth.
+  4 presets: saxophone, alto_sax, tenor_sax, bari_sax
+- **Part.roll()** — rapid repeated notes with velocity ramp for crescendo/
+  decrescendo rolls on any instrument
+- **Vibrato tuning** — all instruments reduced to 0.001 depth for cleaner
+  ensemble sound
+- **Granular synthesis** — grain cloud engine with scatter, pitch
+  variation, and Hanning-windowed grains. Two presets: granular_pad,
+  granular_texture.
+- 30 synth waveforms, 838 tests
+
 ## 0.34.0
 
 - **16 dedicated instrument synths** — physical modeling and specialized
@@ -20,6 +44,8 @@ All notable changes to PyTheory are documented here.
   with 22 new drum patterns
 - **Piano improvements:** brightness scales with pitch, two-stage decay,
   hammer impact with felt character
+- **Vibrato tuning:** reduced across flute, oboe, trumpet, cello for
+  smoother ensemble sound
 - 27 synth waveforms, 10 envelopes, 40+ instrument presets, 80+ drum patterns
 
 ## 0.33.1

docs/guide/drums.rst

@@ -10,7 +10,7 @@ the genre -- they tell the listener's body how to move before a single
 melodic note is played.
 
 PyTheory includes a complete drum system -- 27 synthesized percussion
-sounds, 58 pattern presets across dozens of genres, and 21 fill presets.
+sounds, 80+ pattern presets across dozens of genres, and 21 fill presets.
 Every sound is generated from waveforms; no samples needed.
 
 Drum Sounds
@@ -145,8 +145,8 @@ Each sound has a dedicated synthesizer:
 Pattern Presets
 ---------------
 
-58 patterns spanning genres from rock to Afro-Cuban to electronic.
-Load them with ``Pattern.preset()``:
+80+ patterns spanning genres from rock to Afro-Cuban to electronic to
+world percussion. Load them with ``Pattern.preset()``:
 
 .. code-block:: pycon
 
@@ -193,9 +193,16 @@ adds syncopation.
 rattling hi-hats of trap, the breakneck tempo of drum and bass. These
 patterns were born in drum machines and they still live there.
 
-**Metal/Punk:** metal, blast beat, punk -- Speed and aggression.
-The blast beat is both feet and both hands going as fast as humanly
-possible. Punk strips everything to its essentials.
+**Metal/Punk:** metal, blast beat, punk, double kick, metal blast,
+metal groove, metal gallop -- Speed and aggression. The blast beat is
+both feet and both hands going as fast as humanly possible. Punk strips
+everything to its essentials. The metal kit adds 3 dedicated sounds
+(double kick, china cymbal, stack) and 4 patterns for extreme metal
+subgenres.
+
+**World Percussion:** tabla, dhol, dholak, mridangam, djembe -- Deep
+traditions from across the globe, each with authentic sound sets and
+idiomatic patterns. See the World Percussion section below for details.
 
 **Other:** funk, hip hop, bo diddley, second line, new orleans, waltz,
 12/8 blues, country, gospel, flamenco -- Everything else. The syncopated
@@ -304,6 +311,128 @@ drum pattern and all named parts are mixed together by ``play_score()``:
 
    play_score(score)
 
+World Percussion
+----------------
+
+PyTheory includes dedicated sound sets and pattern presets for
+traditional percussion instruments from around the world. Each
+instrument has its own synthesized sounds that capture the timbral
+character of the real instrument, plus idiomatic rhythmic patterns
+drawn from their musical traditions.
+
+Tabla
+~~~~~
+
+The tabla is a pair of hand drums from the Indian subcontinent -- the
+smaller, higher-pitched *dayan* and the larger, bass *bayan*. It is
+the rhythmic backbone of Hindustani classical music, and one of the
+most expressive percussion instruments ever created. A single tabla
+player can produce an astonishing range of tones by varying finger
+placement, pressure, and striking technique.
+
+**6 sounds** -- covering the primary tabla strokes (na, tin, tun, ge,
+ke, and ti-ra-ki-ta combinations).
+
+**7 patterns:** teental (16 beats, the most common taal), jhaptaal
+(10 beats), rupak (7 beats), dadra (6 beats), keherwa (8 beats, folk
+and light classical), tabla solo, and tiri kita (fast ornamental
+pattern).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=80)
+   score.drums("teental", repeats=4)
+
+Dhol
+~~~~
+
+The dhol is a double-headed barrel drum from Punjab, played with
+sticks. It is the driving force behind bhangra music -- loud,
+energetic, and physically impossible to sit still to.
+
+**3 sounds** -- bass stroke, treble stroke, and rimshot.
+
+**2 patterns:** bhangra (the classic bhangra groove) and dhol chaal
+(a processional rhythm).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=160)
+   score.drums("bhangra", repeats=4)
+
+Dholak
+~~~~~~
+
+The dholak is a smaller, lighter two-headed drum used across South
+Asia in folk music, qawwali, and Bollywood. Played with bare hands,
+it produces a warm, melodic tone.
+
+**3 sounds** -- bass, treble, and slap.
+
+**2 patterns:** qawwali (the rhythmic foundation of Sufi devotional
+music) and dholak folk (a general folk groove).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=120)
+   score.drums("qawwali", repeats=4)
+
+Mridangam
+~~~~~~~~~
+
+The mridangam is a double-headed drum from South India, the
+rhythmic anchor of Carnatic classical music. Its tuning system is
+extraordinarily precise, and its rhythmic vocabulary is among the
+most mathematically complex in the world.
+
+**4 sounds** -- tha, thom, nam, and din.
+
+**2 patterns:** adi talam (the most common Carnatic talam, 8 beats)
+and mridangam korvai (a rhythmic cadence pattern).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=90)
+   score.drums("adi talam", repeats=4)
+
+Djembe
+~~~~~~
+
+The djembe is a rope-tuned goblet drum from West Africa, capable of
+producing a wide range of tones from deep bass to sharp slaps. It is
+central to the drum ensemble traditions of Mali, Guinea, and Senegal.
+
+**3 sounds** -- bass (open center strike), tone (edge strike), and
+slap (sharp edge strike).
+
+**3 patterns:** djembe (a basic accompanying rhythm), kuku (a
+traditional rhythm from Guinea associated with fishing), and soli (a
+solo/celebration rhythm).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=120)
+   score.drums("djembe", repeats=4)
+
+Metal Kit
+~~~~~~~~~
+
+A dedicated percussion kit for extreme metal subgenres, with
+specialized sounds and patterns that go beyond the standard drum kit.
+
+**3 sounds** -- double kick (triggered, tight attack), china cymbal,
+and stack (a short, trashy cymbal choke).
+
+**4 patterns:** double kick (relentless double bass drum pattern),
+metal blast (blast beat with china cymbal accents), metal groove (a
+half-time groove with double kick fills), and metal gallop (the
+classic triplet-feel gallop rhythm).
+
+.. code-block:: python
+
+   score = Score("4/4", bpm=200)
+   score.drums("metal blast", repeats=4)
+
 MIDI Export
 -----------
 

docs/guide/effects.rst

@@ -32,8 +32,8 @@ It's a well-tested order that sounds good by default.
 
 Effects are applied in this fixed order::
 
-    Signal --> Saturation --> Tremolo --> Distortion --> Chorus --> Phaser
-          --> Highpass --> Lowpass --> Delay --> Reverb --> Mix
+    Signal --> Saturation --> Tremolo --> Distortion --> Cabinet --> Chorus
+          --> Phaser --> Highpass --> Lowpass --> Delay --> Reverb --> Mix
 
 Additionally, these per-note effects are applied before the part effects chain:
 
@@ -47,11 +47,12 @@ Part-level effects:
 - **Saturation** first: subtle even-harmonic warmth (tape/tube color).
 - **Tremolo** second: amplitude LFO modulation.
 - **Distortion** third: drives the signal before filtering.
-- **Chorus** fourth: thickens the signal.
-- **Phaser** fifth: swept allpass notches.
-- **Highpass** sixth: removes low-frequency mud.
-- **Lowpass** seventh: shapes the tone (like a tone knob on an amp).
-- **Delay** eighth: echoes the shaped signal (tap delay / tape echo).
+- **Cabinet** fourth: speaker cab simulation (rolloff + presence bump).
+- **Chorus** fifth: thickens the signal.
+- **Phaser** sixth: swept allpass notches.
+- **Highpass** seventh: removes low-frequency mud.
+- **Lowpass** eighth: shapes the tone (like a tone knob on an amp).
+- **Delay** ninth: echoes the shaped signal (tap delay / tape echo).
 - **Reverb** last: places everything in a space (room / hall).
 
 Distortion
@@ -96,6 +97,89 @@ Parameters:
        distortion_drive=10.0,
    )
 
+Cabinet Simulation
+------------------
+
+A real guitar amp doesn't just distort the signal -- the speaker
+cabinet shapes the tone dramatically. A 12-inch speaker in a closed
+cabinet rolls off the harsh high frequencies above 5 kHz and adds a
+presence bump around 2--3 kHz that gives the sound its "in the room"
+quality. Without a cabinet, distortion sounds thin and fizzy. With
+one, it sounds like a real amp.
+
+PyTheory's cabinet simulation applies a speaker rolloff curve (lowpass
+at ~5 kHz) combined with a presence resonance bump, placed in the
+signal chain immediately after distortion -- exactly where it sits in
+a real amp.
+
+Parameters:
+
+- ``cabinet``: Wet/dry mix, 0.0--1.0 (default 0, off).
+
+  - 0.3--0.5 = subtle speaker coloring
+  - 0.6--0.8 = classic amp-in-a-room
+  - 1.0 = full cabinet, no dry signal
+
+.. code-block:: python
+
+   # Classic rock amp tone: distortion into cabinet
+   guitar = score.part(
+       "guitar",
+       synth="saw",
+       envelope="pluck",
+       distortion=0.6,
+       distortion_drive=5.0,
+       cabinet=0.8,
+   )
+
+   # Clean amp with just cabinet warmth (no distortion)
+   clean = score.part(
+       "clean",
+       synth="triangle",
+       envelope="pluck",
+       cabinet=0.5,
+   )
+
+Analog Drift
+------------
+
+Real analog synthesizers are never perfectly in tune. The voltage-
+controlled oscillators drift slightly over time as components warm up
+and temperature fluctuates. This imperfection is actually a big part
+of why vintage analog synths sound so appealing -- the subtle pitch
+wandering gives each note a unique, living quality that static digital
+oscillators lack.
+
+The ``analog_drift`` parameter adds slow, random pitch variation to
+each oscillator, modeling this vintage behavior.
+
+Parameters:
+
+- ``analog_drift``: Drift amount, 0.0--1.0 (default 0, off).
+
+  - 0.05--0.1 = subtle warmth (studio-grade analog)
+  - 0.15--0.25 = noticeable drift (vintage gear warming up)
+  - 0.3+ = unstable, wobbly (broken tape machine)
+
+.. code-block:: python
+
+   # Warm vintage pad
+   pad = score.part(
+       "pad",
+       synth="supersaw",
+       envelope="pad",
+       analog_drift=0.1,
+       chorus=0.3,
+   )
+
+   # Lo-fi detuned lead
+   lead = score.part(
+       "lead",
+       synth="saw",
+       envelope="pluck",
+       analog_drift=0.25,
+   )
+
 Chorus
 ------
 

docs/guide/sequencing.rst

@@ -574,3 +574,107 @@ Define sections with ``score.section()`` and repeat them with
 Use any names you want — ``"intro"``, ``"verse"``, ``"chorus"``,
 ``"bridge"``, ``"drop"``, ``"breakdown"``, ``"outro"``, or anything
 that makes sense for your song. The names are just labels.
+
+Guitar Strumming
+----------------
+
+Any part with a fretboard can strum chords using real fingering
+positions. The ``strum()`` method looks up the chord on the fretboard,
+gets the correct voicing, and plays all strings as a chord.
+
+.. code-block:: python
+
+   from pytheory import Fretboard
+
+   guitar = score.part("guitar", instrument="acoustic_guitar",
+                       fretboard=Fretboard.guitar())
+
+   guitar.strum("Am", Duration.HALF, direction="down")
+   guitar.strum("G", Duration.HALF, direction="up")
+   guitar.strum("F", Duration.WHOLE)
+
+Works with any fretboard instrument — guitar, ukulele, banjo, mandolin.
+Works with any guitar preset — clean, crunch, distorted, orange, metal.
+
+Pitch Bends
+-----------
+
+Bend a note's pitch up or down over its duration. Essential for guitar
+bends, sitar meends, trombone slides, and vocal-style expression.
+
+.. code-block:: python
+
+   # Guitar bend: D up to E (2 semitones)
+   guitar.add("D4", Duration.HALF, bend=2, bend_type="smooth")
+
+   # Release bend: E back down to D
+   guitar.add("E4", Duration.HALF, bend=-2)
+
+   # Blues curl: hold then bend at the end
+   guitar.add("C4", Duration.HALF, bend=1, bend_type="late")
+
+Three bend types:
+
+- ``"smooth"`` — logarithmic (default). Perceptually even pitch change.
+- ``"linear"`` — linear frequency interpolation. Mechanical/synth feel.
+- ``"late"`` — holds the starting pitch for 60%, bends in the last 40%.
+  The classic blues "curl."
+
+Rolls
+-----
+
+Rapid repeated notes with a velocity ramp — perfect for timpani
+rolls, snare rolls, tremolo on any instrument. The velocity ramps
+from ``velocity_start`` to ``velocity_end`` for crescendo or
+decrescendo effects.
+
+.. code-block:: python
+
+   # Timpani crescendo roll
+   timp = score.part("timp", instrument="timpani")
+   timp.roll("C3", Duration.WHOLE, velocity_start=20, velocity_end=110)
+   timp.add("C3", Duration.HALF, velocity=127)  # big accent
+
+   # Snare roll with 32nd notes
+   snare = score.part("snare", synth="noise", envelope="pluck")
+   snare.roll("C4", Duration.HALF, speed=0.125,
+              velocity_start=40, velocity_end=100)
+
+   # Decrescendo (loud to quiet)
+   timp.roll("G2", Duration.WHOLE, velocity_start=100, velocity_end=30)
+
+Parameters:
+
+- ``velocity_start``: Starting velocity (default 40).
+- ``velocity_end``: Ending velocity (default 100).
+- ``speed``: Note subdivision (default ``Duration.SIXTEENTH``).
+  Use ``0.125`` for 32nd notes, ``Duration.EIGHTH`` for 8th notes.
+
+Tuning Systems
+--------------
+
+A Score can use any tuning system and temperament:
+
+.. code-block:: python
+
+   # Baroque harpsichord — meantone tuning, A=415
+   score = Score("4/4", bpm=80, temperament="meantone",
+                 reference_pitch=415.0)
+
+   # Indian classical — 22-shruti system
+   score = Score("4/4", bpm=75, system="shruti")
+
+   # Just intonation — pure intervals
+   score = Score("4/4", bpm=90, temperament="just")
+
+Temperaments: ``"equal"`` (default), ``"pythagorean"``, ``"meantone"``,
+``"just"``.
+
+Custom equal temperaments via the ``TET()`` factory:
+
+.. code-block:: python
+
+   from pytheory import TET
+
+   edo19 = TET(19)  # 19-tone equal temperament
+   score = Score("4/4", bpm=100, system=edo19)

docs/guide/synths.rst

@@ -1,7 +1,7 @@
 Synthesizers
 ============
 
-PyTheory includes 13 built-in waveforms and 10 ADSR envelope presets.
+PyTheory includes 30 built-in waveforms and 10 ADSR envelope presets.
 Every sound is generated from scratch -- no samples or external audio
 files needed.
 
@@ -233,7 +233,7 @@ shapes the amplitude over time for natural-sounding notes:
 - **Sustain** -- the held volume while the note is on.
 - **Release** -- how quickly it fades to silence after the note ends.
 
-PyTheory includes 8 presets:
+PyTheory includes 10 presets:
 
 .. code-block:: python
 
@@ -386,11 +386,235 @@ more "wooden" than a raw saw wave.
 
    violin = score.part("violin", synth="strings_synth")
 
+Dedicated Instrument Synths
+--------------------------
+
+Beyond the classic and physical modeling waveforms, PyTheory includes
+17 dedicated instrument synths. Each one uses tailored synthesis
+techniques -- additive harmonics, formant shaping, body resonance
+modeling, and specialized envelopes -- to capture the character of a
+specific acoustic instrument. These are the waveforms that bring the
+total count to 30.
+
+Piano Synth
+~~~~~~~~~~~
+
+Hammer-strike envelope with body resonance and subtle inharmonicity.
+Models the way a felt hammer excites steel strings inside a wooden
+soundboard.
+
+.. code-block:: python
+
+   piano = score.part("piano", synth="piano_synth")
+
+Bass Guitar Synth
+~~~~~~~~~~~~~~~~~
+
+Plucked string model with finger-damped harmonics and low-end warmth.
+
+.. code-block:: python
+
+   bass = score.part("bass", synth="bass_guitar_synth")
+
+Flute Synth
+~~~~~~~~~~~~
+
+Breathy noise excitation through a resonant tube model, with
+overblowing behavior at higher velocities.
+
+.. code-block:: python
+
+   flute = score.part("flute", synth="flute_synth")
+
+Trumpet Synth
+~~~~~~~~~~~~~
+
+Brass lip-buzz model with spectral brightness that increases with
+velocity, plus a characteristic brassy edge from shaped harmonics.
+
+.. code-block:: python
+
+   trumpet = score.part("trumpet", synth="trumpet_synth")
+
+Clarinet Synth
+~~~~~~~~~~~~~~
+
+Cylindrical bore model producing mostly odd harmonics, giving the
+characteristic hollow, woody tone.
+
+.. code-block:: python
+
+   clarinet = score.part("clarinet", synth="clarinet_synth")
+
+Oboe Synth
+~~~~~~~~~~~
+
+Double-reed model with nasal formant shaping and a buzzy, penetrating
+timbre.
+
+.. code-block:: python
+
+   oboe = score.part("oboe", synth="oboe_synth")
+
+Marimba Synth
+~~~~~~~~~~~~~
+
+Tuned bar model with a soft mallet attack and a warm, resonant decay
+that emphasizes the fundamental.
+
+.. code-block:: python
+
+   marimba = score.part("marimba", synth="marimba_synth")
+
+Harpsichord Synth
+~~~~~~~~~~~~~~~~~
+
+Plucked-string model with a bright, immediate attack and rapid decay
+-- the characteristic "plink" of a quill plucking a string.
+
+.. code-block:: python
+
+   harpsi = score.part("harpsi", synth="harpsichord_synth")
+
+Cello Synth
+~~~~~~~~~~~
+
+Bowed string model with body formants at cello resonance frequencies,
+producing a rich, warm, sustained tone.
+
+.. code-block:: python
+
+   cello = score.part("cello", synth="cello_synth")
+
+Harp Synth
+~~~~~~~~~~
+
+Plucked string with longer sustain and gentle high-frequency rolloff,
+modeling nylon strings on a resonant frame.
+
+.. code-block:: python
+
+   harp = score.part("harp", synth="harp_synth")
+
+Upright Bass Synth
+~~~~~~~~~~~~~~~~~~
+
+Pizzicato double bass with woody body resonance and a thumpy low end.
+
+.. code-block:: python
+
+   bass = score.part("bass", synth="upright_bass_synth")
+
+Acoustic Guitar Synth
+~~~~~~~~~~~~~~~~~~~~~
+
+Steel-string model with pick transient, body resonance, and natural
+string decay.
+
+.. code-block:: python
+
+   guitar = score.part("guitar", synth="acoustic_guitar_synth")
+
+Electric Guitar Synth
+~~~~~~~~~~~~~~~~~~~~~
+
+Magnetic pickup model with brighter harmonics and less body resonance
+than the acoustic, ready for effects processing.
+
+.. code-block:: python
+
+   eguitar = score.part("eguitar", synth="electric_guitar_synth")
+
+Sitar Synth
+~~~~~~~~~~~~
+
+Sympathetic string resonance with the characteristic buzzy "jawari"
+bridge, producing a shimmering, metallic sustain.
+
+.. code-block:: python
+
+   sitar = score.part("sitar", synth="sitar_synth")
+
+Timpani Synth
+~~~~~~~~~~~~~
+
+Large kettle drum with definite pitch. Inharmonic membrane modes
+(1.0, 1.5, 1.99, 2.44), felt mallet attack, copper kettle resonance.
+Use ``Part.roll()`` for crescendo timpani rolls.
+
+.. code-block:: python
+
+   timp = score.part("timp", synth="timpani_synth")
+   timp.roll("C3", Duration.WHOLE, velocity_start=20, velocity_end=110)
+
+Saxophone Synth
+~~~~~~~~~~~~~~~
+
+Single reed through a conical brass bore. All harmonics with strong
+mids, reed buzz, and brass body warmth. Four presets: ``saxophone``,
+``alto_sax``, ``tenor_sax``, ``bari_sax``.
+
+.. code-block:: python
+
+   sax = score.part("sax", instrument="tenor_sax")
+
+Granular Synth
+~~~~~~~~~~~~~~
+
+Grain cloud synthesis — chops a source waveform into tiny overlapping
+grains (10-200ms), each windowed and optionally pitch/time scattered.
+Creates textures impossible with other synthesis: frozen tones,
+shimmering clouds, evolving pads, glitchy stutters.
+
+.. code-block:: python
+
+   # Atmospheric granular pad
+   pad = score.part("pad", instrument="granular_pad")
+
+   # Granular with filter envelope sweep + resonance
+   texture = score.part("texture", synth="granular_synth", envelope="pad",
+                        filter_amount=4000, filter_attack=0.5,
+                        filter_decay=1.5, filter_sustain=0.3,
+                        lowpass=600, lowpass_q=3.0,
+                        reverb=0.5, reverb_type="taj_mahal")
+
+Parameters (passed as synth kwargs):
+
+- ``grain_size``: Duration per grain in seconds (default 0.04).
+- ``density``: Grains per second (default 50). Higher = denser cloud.
+- ``scatter``: Random position jitter 0-1 (default 0.5).
+- ``pitch_var``: Per-grain pitch randomization in cents (default 12).
+- ``source``: Base waveform — ``"saw"``, ``"sine"``, ``"triangle"``,
+  ``"square"``, ``"noise"``.
+
+Analog Oscillator Drift
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+All waveform synths support the ``analog_drift`` parameter, which adds
+subtle, slow random pitch variation to each oscillator -- modeling the
+voltage instability of vintage analog circuits. This is what makes a
+real Minimoog sound slightly different on every note, and why analog
+synths feel "alive" compared to their digital counterparts.
+
+.. code-block:: python
+
+   # Subtle vintage drift
+   pad = score.part("pad", synth="saw", analog_drift=0.1)
+
+   # More pronounced, wobbly analog character
+   lead = score.part("lead", synth="square", analog_drift=0.3)
+
+Drift values:
+
+- **0.05--0.1** = subtle warmth (studio-grade analog)
+- **0.15--0.25** = noticeable drift (vintage gear warming up)
+- **0.3+** = unstable, wobbly (broken tape machine)
+
 Instrument Presets
 ------------------
 
 Instead of choosing synth + envelope + effects manually, use an
-instrument preset — 38 predefined combinations that approximate real
+instrument preset — 40+ predefined combinations that approximate real
 instruments:
 
 .. code-block:: python

docs/index.rst

@@ -77,15 +77,18 @@ What's Inside
   numbers), scale recommendation, modulation, voice leading
 - **Sequencing** — Score, Parts, arpeggiator, legato/glide, velocity,
   swing, humanize, tempo changes, song sections with repeat
-- **Synthesis** — 13 waveforms (including Karplus-Strong pluck, Hammond organ,
-  bowed string), 10 envelopes, 38 instrument presets, configurable FM,
-  sub-oscillator, noise layer, filter envelope, velocity-to-brightness,
-  detune, stereo pan/spread, 58 drum patterns (stereo panned), 21 fills
+- **Synthesis** — 30 waveforms (including Karplus-Strong pluck, Hammond organ,
+  bowed string, and 14 dedicated instrument synths), 10 envelopes, 40+
+  instrument presets, configurable FM, sub-oscillator, noise layer, filter
+  envelope, velocity-to-brightness, analog oscillator drift, detune, stereo
+  pan/spread, strumming, 80+ drum patterns (stereo panned, including world
+  percussion), 21 fills
 - **Effects** — reverb (algorithmic + 7 convolution IRs, stereo), delay,
-  lowpass/highpass (with resonance), distortion, saturation, chorus,
-  phaser, tremolo, sidechain compression, automation, LFOs. Master bus
-  compressor/limiter
-- **Instruments** — 25 presets with fingering generation
+  lowpass/highpass (with resonance), distortion, cabinet simulation,
+  saturation, chorus, phaser, tremolo, analog drift, sidechain compression,
+  automation, LFOs. Master bus compressor/limiter
+- **Instruments** — 49 presets with fingering generation, guitar strumming,
+  pitch bends
 - **Output** — stereo playback, WAV export, MIDI import/export
 - **Interface** — REPL with tab completion, CLI (15 commands), ``pytheory demo``
 - **AI-friendly** — Claude Code can compose

examples/songs.py

@@ -15,7 +15,8 @@ Usage:
 
 import sounddevice as sd
 
-from pytheory import Chord, Key, Pattern, Duration, Score
+from pytheory import Chord, Key, Pattern, Duration, Score, Tone, TonedScale, SYSTEMS
+from pytheory.rhythm import DrumSound, _Hit
 from pytheory.play import render_score, SAMPLE_RATE
 
 
@@ -45,18 +46,15 @@ def bossa_nova_girl():
     score = Score("4/4", bpm=140)
     score.drums("bossa nova", repeats=4)
 
-    rhodes = score.part("rhodes", synth="fm", envelope="piano",
+    rhodes = score.part("rhodes", instrument="electric_piano",
                         volume=0.3, pan=-0.3,
-                        reverb=0.4, reverb_decay=1.8, reverb_type="plate",
-                        detune=8, humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="pluck",
+                        reverb=0.4, reverb_decay=1.8, reverb_type="plate")
+    lead = score.part("lead", instrument="flute",
                       volume=0.45, pan=0.3,
                       delay=0.25, delay_time=0.32, delay_feedback=0.35,
-                      reverb=0.2, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.45, pan=0.0, lowpass=600,
-                      humanize=0.15)
+                      reverb=0.2, reverb_type="plate")
+    bass = score.part("bass", instrument="upright_bass",
+                      volume=0.45, pan=0.0, lowpass=600)
 
     for sym in ["Am", "Am", "Dm", "Dm", "E7", "E7", "Am", "Am"]:
         rhodes.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -90,19 +88,16 @@ def bebop_in_bb():
     score = Score("4/4", bpm=160)
     score.drums("bebop", repeats=8, fill="jazz", fill_every=8)
 
-    rhodes = score.part("rhodes", synth="fm", envelope="piano",
+    rhodes = score.part("rhodes", instrument="electric_piano",
                         volume=0.25, pan=-0.3,
-                        reverb=0.35, reverb_decay=1.2, reverb_type="plate",
-                        detune=8, humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="pluck",
+                        reverb=0.35, reverb_decay=1.2, reverb_type="plate")
+    lead = score.part("lead", instrument="trumpet",
                       volume=0.4, pan=0.25,
                       lowpass=4000, lowpass_q=1.1,
                       delay=0.15, delay_time=0.19, delay_feedback=0.25,
-                      reverb=0.15, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="triangle", envelope="pluck",
-                      volume=0.4, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      reverb=0.15, reverb_type="plate")
+    bass = score.part("bass", instrument="upright_bass",
+                      volume=0.4, pan=0.0, lowpass=500)
 
     for sym in ["Bb", "Gm", "Cm", "F7"] * 2:
         rhodes.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -148,14 +143,12 @@ def salsa_descarga():
                       volume=0.2, pan=-0.35,
                       reverb=0.3, reverb_type="plate", lowpass=2000,
                       detune=10, humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="pluck",
+    lead = score.part("lead", instrument="trumpet",
                       volume=0.4, pan=0.3,
                       delay=0.2, delay_time=0.167, delay_feedback=0.3,
-                      reverb=0.15, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="pulse", envelope="pluck",
-                      volume=0.45, pan=0.0, lowpass=500, lowpass_q=1.3,
-                      humanize=0.15)
+                      reverb=0.15, reverb_type="plate")
+    bass = score.part("bass", instrument="synth_bass",
+                      volume=0.45, pan=0.0, lowpass=500, lowpass_q=1.3)
 
     for sym in ["Em7b5", "A7", "Dm7", "Bbmaj7"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -191,20 +184,17 @@ def afrobeat_groove():
     score = Score("4/4", bpm=115)
     score.drums("afrobeat", repeats=8, fill="afrobeat", fill_every=8)
 
-    pads = score.part("pads", synth="supersaw", envelope="pad",
+    pads = score.part("pads", instrument="synth_pad",
                       volume=0.2, pan=-0.3,
                       reverb=0.4, reverb_decay=2.0, reverb_type="cathedral",
-                      lowpass=3000, detune=10, spread=0.6,
-                      humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="pluck",
+                      lowpass=3000)
+    lead = score.part("lead", instrument="trumpet",
                       volume=0.4, pan=0.3,
                       lowpass=3000, lowpass_q=1.0,
                       delay=0.2, delay_time=0.26, delay_feedback=0.3,
-                      reverb=0.15, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.5, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      reverb=0.15, reverb_type="plate")
+    bass = score.part("bass", instrument="bass_guitar",
+                      volume=0.5, pan=0.0, lowpass=500)
 
     for sym in ["Em", "Am", "D", "C"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -241,14 +231,12 @@ def reggae_one_drop():
                         reverb=0.5, reverb_decay=2.0, reverb_type="cathedral",
                         lowpass=2000, detune=8,
                         humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="strings",
+    lead = score.part("lead", instrument="flute",
                       volume=0.4, pan=0.3,
                       delay=0.35, delay_time=0.5625, delay_feedback=0.45,
-                      reverb=0.3, reverb_type="cathedral",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.55, pan=0.0, lowpass=400, lowpass_q=1.3,
-                      humanize=0.15)
+                      reverb=0.3, reverb_type="cathedral")
+    bass = score.part("bass", instrument="bass_guitar",
+                      volume=0.55, pan=0.0, lowpass=400, lowpass_q=1.3)
 
     for sym in ["G", "C", "D", "C"] * 2:
         chords.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -286,15 +274,13 @@ def funk_workout():
                         volume=0.25, pan=-0.4,
                         lowpass=2500, reverb=0.15, reverb_type="plate",
                         sidechain=0.4, humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="pluck",
+    lead = score.part("lead", instrument="synth_lead",
                       volume=0.4, pan=0.35,
                       lowpass=3500, lowpass_q=1.5,
                       delay=0.15, delay_time=0.15, delay_feedback=0.25,
-                      reverb=0.1, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="pulse", envelope="pluck",
-                      volume=0.5, pan=0.0, lowpass=600, lowpass_q=1.2,
-                      humanize=0.15)
+                      reverb=0.1, reverb_type="plate")
+    bass = score.part("bass", instrument="synth_bass",
+                      volume=0.5, pan=0.0, lowpass=600, lowpass_q=1.2)
 
     for sym in ["Em", "Am", "D", "B7"] * 2:
         chords.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -332,18 +318,16 @@ def blues_shuffle():
     score = Score("12/8", bpm=70)
     score.drums("12/8 blues", repeats=6)
 
-    chords = score.part("chords", synth="fm", envelope="piano",
+    chords = score.part("chords", instrument="electric_piano",
                         volume=0.3, pan=-0.3,
-                        reverb=0.3, reverb_decay=1.5, reverb_type="plate",
-                        detune=8, humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="pluck",
+                        reverb=0.3, reverb_decay=1.5, reverb_type="plate")
+    lead = score.part("lead", instrument="trumpet",
                       volume=0.45, pan=0.25,
                       reverb=0.3, reverb_decay=1.2, reverb_type="plate",
                       delay=0.2, delay_time=0.43, delay_feedback=0.3,
-                      lowpass=3500, humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.5, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      lowpass=3500)
+    bass = score.part("bass", instrument="upright_bass",
+                      volume=0.5, pan=0.0, lowpass=500)
 
     for sym in ["A", "A", "D", "D", "E7", "A"]:
         chords.add(Chord.from_symbol(sym), Duration.DOTTED_HALF)
@@ -381,19 +365,16 @@ def samba_de_janeiro():
     score = Score("4/4", bpm=170)
     score.drums("samba", repeats=8, fill="samba", fill_every=8)
 
-    pads = score.part("pads", synth="supersaw", envelope="pad",
+    pads = score.part("pads", instrument="synth_pad",
                       volume=0.2, pan=-0.3,
                       reverb=0.45, reverb_decay=2.0, reverb_type="plate",
-                      lowpass=4000, detune=10, spread=0.5,
-                      humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="pluck",
+                      lowpass=4000)
+    lead = score.part("lead", instrument="flute",
                       volume=0.45, pan=0.3,
                       delay=0.2, delay_time=0.176, delay_feedback=0.3,
-                      reverb=0.15, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.45, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      reverb=0.15, reverb_type="plate")
+    bass = score.part("bass", instrument="bass_guitar",
+                      volume=0.45, pan=0.0, lowpass=500)
 
     for sym in ["G", "Em", "Am", "D7"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -426,18 +407,15 @@ def jazz_waltz():
     score = Score("3/4", bpm=150)
     score.drums("waltz", repeats=16)
 
-    rhodes = score.part("rhodes", synth="fm", envelope="piano",
+    rhodes = score.part("rhodes", instrument="electric_piano",
                         volume=0.3, pan=-0.3,
-                        reverb=0.4, reverb_decay=2.0, reverb_type="cathedral",
-                        detune=8, humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="strings",
+                        reverb=0.4, reverb_decay=2.0, reverb_type="cathedral")
+    lead = score.part("lead", instrument="flute",
                       volume=0.4, pan=0.25,
                       reverb=0.3, reverb_decay=1.5, reverb_type="plate",
-                      delay=0.2, delay_time=0.4, delay_feedback=0.3,
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.4, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      delay=0.2, delay_time=0.4, delay_feedback=0.3)
+    bass = score.part("bass", instrument="upright_bass",
+                      volume=0.4, pan=0.0, lowpass=500)
 
     for _ in range(2):
         for sym in ["Fmaj7", "Gm", "C7", "Fmaj7"]:
@@ -471,20 +449,19 @@ def house_anthem():
     score = Score("4/4", bpm=124)
     score.drums("house", repeats=8, fill="house", fill_every=8)
 
-    pads = score.part("pads", synth="supersaw", envelope="pad",
+    pads = score.part("pads", instrument="synth_pad",
                       volume=0.25, pan=-0.3,
                       reverb=0.5, reverb_decay=2.5, reverb_type="cathedral",
-                      lowpass=5000, detune=12, spread=0.7,
-                      sidechain=0.6, humanize=0.2)
-    lead = score.part("lead", synth="saw", envelope="staccato",
+                      lowpass=5000,
+                      sidechain=0.6)
+    lead = score.part("lead", instrument="synth_lead",
                       volume=0.35, pan=0.3,
                       lowpass=2000, lowpass_q=2.0,
                       delay=0.2, delay_time=0.242, delay_feedback=0.35,
-                      reverb=0.15, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.55, pan=0.0, lowpass=300,
-                      sidechain=0.5, humanize=0.15)
+                      reverb=0.15, reverb_type="plate")
+    bass = score.part("bass", instrument="808_bass",
+                      volume=0.55, pan=0.0,
+                      sidechain=0.5)
 
     for sym in ["Cm", "Ab", "Bb", "Cm"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -538,14 +515,12 @@ def dub_kingston():
                         reverb=0.6, reverb_decay=2.5, reverb_type="cathedral",
                         lowpass=1500, lowpass_q=0.9, detune=8,
                         humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="strings",
+    lead = score.part("lead", instrument="flute",
                       volume=0.4, pan=0.3,
                       delay=0.45, delay_time=0.625, delay_feedback=0.5,
-                      reverb=0.35, reverb_decay=2.0, reverb_type="cathedral",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.6, pan=0.0, lowpass=400, lowpass_q=1.5,
-                      humanize=0.15)
+                      reverb=0.35, reverb_decay=2.0, reverb_type="cathedral")
+    bass = score.part("bass", instrument="bass_guitar",
+                      volume=0.6, pan=0.0, lowpass=400, lowpass_q=1.5)
     siren = score.part("siren", synth="pwm_slow", envelope="pad",
                        volume=0.15, pan=0.5,
                        reverb=0.7, reverb_decay=3.0, reverb_type="cathedral",
@@ -585,20 +560,20 @@ def techno_minimal():
     score = Score("4/4", bpm=130)
     score.drums("techno", repeats=8, fill="house", fill_every=8)
 
-    pad = score.part("pad", synth="supersaw", envelope="pad",
+    pad = score.part("pad", instrument="synth_pad",
                      volume=0.2, pan=-0.3,
                      reverb=0.5, reverb_decay=3.0, reverb_type="cathedral",
-                     lowpass=3000, detune=12, spread=0.7,
-                     sidechain=0.6, humanize=0.2)
+                     lowpass=3000,
+                     sidechain=0.6)
     lead = score.part("lead", synth="pwm_fast", envelope="staccato",
                       volume=0.35, pan=0.3,
                       lowpass=1500, lowpass_q=3.0,
                       delay=0.3, delay_time=0.231, delay_feedback=0.4,
                       reverb=0.1, reverb_type="plate",
                       humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.55, pan=0.0, lowpass=250,
-                      sidechain=0.5, humanize=0.15)
+    bass = score.part("bass", instrument="808_bass",
+                      volume=0.55, pan=0.0,
+                      sidechain=0.5)
 
     for sym in ["Fm", "Db", "Eb", "Fm"] * 2:
         pad.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -624,18 +599,15 @@ def gospel_shuffle():
     score = Score("4/4", bpm=108)
     score.drums("gospel", repeats=8, fill="buildup", fill_every=8)
 
-    organ = score.part("organ", synth="fm", envelope="organ",
+    organ = score.part("organ", instrument="organ",
                        volume=0.3, pan=-0.3,
-                       reverb=0.45, reverb_decay=2.0, reverb_type="cathedral",
-                       detune=8, humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="pluck",
+                       reverb=0.45, reverb_decay=2.0, reverb_type="cathedral")
+    lead = score.part("lead", instrument="flute",
                       volume=0.4, pan=0.3,
                       delay=0.2, delay_time=0.278, delay_feedback=0.3,
-                      reverb=0.2, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.45, pan=0.0, lowpass=500,
-                      humanize=0.15)
+                      reverb=0.2, reverb_type="plate")
+    bass = score.part("bass", instrument="upright_bass",
+                      volume=0.45, pan=0.0, lowpass=500)
 
     for sym in ["C", "Am", "F", "G"] * 2:
         organ.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -691,20 +663,18 @@ def dub_delay_madness():
                         volume=0.15, pan=-0.4,
                         reverb=0.7, reverb_decay=3.0, reverb_type="cathedral",
                         lowpass=1200, detune=8, humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.6, pan=0.0, lowpass=350, lowpass_q=1.5,
-                      humanize=0.15)
+    bass = score.part("bass", instrument="bass_guitar",
+                      volume=0.6, pan=0.0, lowpass=350, lowpass_q=1.5)
     siren = score.part("siren", synth="pwm_slow", envelope="pad",
                        volume=0.12, pan=0.5,
                        reverb=0.8, reverb_decay=4.0, reverb_type="cathedral",
                        delay=0.4, delay_time=0.88, delay_feedback=0.6,
                        lowpass=900, detune=10)
     # Melodica stabs — sparse, lots of delay
-    melodica = score.part("melodica", synth="triangle", envelope="pluck",
+    melodica = score.part("melodica", instrument="flute",
                           volume=0.35, pan=0.3,
                           delay=0.6, delay_time=0.66, delay_feedback=0.55,
-                          reverb=0.5, reverb_decay=2.5, reverb_type="cathedral",
-                          humanize=0.2)
+                          reverb=0.5, reverb_decay=2.5, reverb_type="cathedral")
 
     for sym in ["Em", "Em", "Am", "Am", "Em", "Em", "Bm", "Em"]:
         chords.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -741,19 +711,16 @@ def drum_and_bass():
     score = Score("4/4", bpm=174)
     score.drums("drum and bass", repeats=8, fill="buildup", fill_every=8)
 
-    pads = score.part("pads", synth="supersaw", envelope="pad",
+    pads = score.part("pads", instrument="synth_pad",
                       volume=0.25, pan=-0.3,
                       reverb=0.5, reverb_decay=2.5, reverb_type="plate",
-                      lowpass=4000, detune=10, spread=0.6,
-                      humanize=0.2)
-    lead = score.part("lead", synth="triangle", envelope="strings",
+                      lowpass=4000)
+    lead = score.part("lead", instrument="flute",
                       volume=0.4, pan=0.3,
                       delay=0.3, delay_time=0.172, delay_feedback=0.4,
-                      reverb=0.25, reverb_type="plate",
-                      humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.55, pan=0.0, lowpass=300,
-                      humanize=0.15)
+                      reverb=0.25, reverb_type="plate")
+    bass = score.part("bass", instrument="808_bass",
+                      volume=0.55, pan=0.0)
 
     for sym in ["Am", "F", "C", "G"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -787,25 +754,24 @@ def drake_vibes():
     score = Score("4/4", bpm=68)
     score.drums("trap", repeats=8, fill="trap", fill_every=8)
 
-    pads = score.part("pads", synth="supersaw", envelope="pad",
+    pads = score.part("pads", instrument="synth_pad",
                       volume=0.2, pan=-0.25,
                       reverb=0.5, reverb_decay=3.0, reverb_type="cathedral",
-                      lowpass=2500, detune=12, spread=0.6,
-                      sidechain=0.4, humanize=0.2)
-    bells = score.part("bells", synth="fm", envelope="bell",
+                      lowpass=2500,
+                      sidechain=0.4)
+    bells = score.part("bells", instrument="vibraphone",
                        volume=0.3, pan=0.4,
                        reverb=0.4, reverb_decay=2.0, reverb_type="plate",
-                       delay=0.25, delay_time=0.44, delay_feedback=0.35,
-                       humanize=0.2)
+                       delay=0.25, delay_time=0.44, delay_feedback=0.35)
     lead = score.part("lead", synth="pwm_slow", envelope="strings",
                       volume=0.35, pan=-0.2,
                       reverb=0.3, reverb_type="cathedral", lowpass=2000,
                       delay=0.2, delay_time=0.88, delay_feedback=0.3,
                       humanize=0.2)
-    bass = score.part("bass", synth="sine", envelope="pluck",
-                      volume=0.6, pan=0.0, lowpass=200, lowpass_q=1.8,
+    bass = score.part("bass", instrument="808_bass",
+                      volume=0.6, pan=0.0,
                       distortion=0.4, distortion_drive=2.0,
-                      sidechain=0.3, humanize=0.15)
+                      sidechain=0.3)
 
     for sym in ["Ebm", "B", "Gb", "Db"] * 2:
         pads.add(Chord.from_symbol(sym), Duration.WHOLE)
@@ -858,14 +824,14 @@ def neon_grid():
     score._drum_pattern_beats = max(score._drum_pattern_beats, 32.0)
 
     sky = score.part(
-        "sky", synth="supersaw", envelope="pad",
+        "sky", instrument="synth_pad",
         volume=0.18, pan=0.0,
         detune=30, spread=1.0,
         reverb=0.6, reverb_decay=3.5,
         chorus=0.3, sidechain=0.5,
     )
     acid_l = score.part(
-        "acid_l", synth="saw", envelope="pad",
+        "acid_l", instrument="acid_bass",
         volume=0.35, pan=-0.7,
         legato=True, glide=0.025,
         distortion=0.9, distortion_drive=12.0,
@@ -875,7 +841,7 @@ def neon_grid():
     acid_l.lfo("lowpass", rate=0.5, min=400, max=3000, bars=8, shape="sine")
 
     acid_r = score.part(
-        "acid_r", synth="saw", envelope="pad",
+        "acid_r", instrument="acid_bass",
         volume=0.3, pan=0.7,
         legato=True, glide=0.02,
         distortion=0.85, distortion_drive=10.0,
@@ -885,7 +851,7 @@ def neon_grid():
     acid_r.lfo("lowpass", rate=0.33, min=500, max=2500, bars=8, shape="triangle")
 
     sub = score.part(
-        "sub", synth="sine", envelope="pluck",
+        "sub", instrument="808_bass",
         volume=0.55, pan=0.0,
         lowpass=160, sidechain=0.85, sidechain_release=0.08,
     )
@@ -896,12 +862,12 @@ def neon_grid():
         detune=15, spread=0.7,
     )
     bell_l = score.part(
-        "bell_l", synth="fm", envelope="bell",
+        "bell_l", instrument="vibraphone",
         volume=0.1, pan=-1.0,
         reverb=0.7, reverb_decay=3.0,
     )
     bell_r = score.part(
-        "bell_r", synth="fm", envelope="bell",
+        "bell_r", instrument="vibraphone",
         volume=0.1, pan=1.0,
         reverb=0.7, reverb_decay=3.0,
         delay=0.2, delay_time=0.8, delay_feedback=0.4,
@@ -1026,14 +992,14 @@ def dance_party():
             score._drum_hits.append(_Hit(DrumSound.KICK, bar * 4.0 + beat, 120))
     score._drum_pattern_beats = max(score._drum_pattern_beats, 32.0)
 
-    bass = score.part("bass", synth="square", envelope="pluck",
+    bass = score.part("bass", instrument="synth_bass",
                       volume=0.45, lowpass=500, lowpass_q=1.3,
                       sidechain=0.75, sidechain_release=0.12)
-    sparkle = score.part("sparkle", synth="fm", envelope="bell",
+    sparkle = score.part("sparkle", instrument="vibraphone",
                          volume=0.3, pan=0.4, reverb=0.3, reverb_decay=1.5,
                          delay=0.2, delay_time=0.234, delay_feedback=0.3)
-    chords_part = score.part("chords", synth="supersaw", envelope="pad",
-                             volume=0.2, detune=12, spread=0.7,
+    chords_part = score.part("chords", instrument="synth_pad",
+                             volume=0.2,
                              reverb=0.4, reverb_type="plate", sidechain=0.7)
     fun = score.part("fun", synth="square", envelope="staccato",
                      volume=0.2, pan=-0.5, delay=0.15, delay_time=0.117,
@@ -1080,14 +1046,13 @@ def temple_bell():
     score = Score("4/4", bpm=65)
     score.drums("bolero", repeats=8)
 
-    koto = score.part("koto", synth="triangle", envelope="pluck",
+    koto = score.part("koto", instrument="koto",
                       volume=0.45, pan=0.2,
-                      reverb=0.5, reverb_type="taj_mahal",
-                      humanize=0.3)
+                      reverb=0.5, reverb_type="taj_mahal")
     drone = score.part("drone", synth="sine", envelope="pad",
                        volume=0.15, reverb=0.6, reverb_type="taj_mahal",
                        chorus=0.15, chorus_rate=0.2)
-    bell = score.part("bell", synth="fm", envelope="bell",
+    bell = score.part("bell", instrument="vibraphone",
                       volume=0.1, pan=-0.6,
                       reverb=0.8, reverb_type="taj_mahal")
 
@@ -1235,6 +1200,357 @@ def greensleeves():
     play_song(score, "Greensleeves — Renaissance Lute (Meantone, A=415)")
 
 
+def tabla_solo_yaman():
+    """Tabla solo with tanpura drone, sitar, and Raga Yaman — 22-shruti tuning."""
+    shruti = SYSTEMS["shruti"]
+    score = Score("4/4", bpm=160, system=shruti)
+    h = _Hit
+
+    NA = DrumSound.TABLA_NA
+    TI = DrumSound.TABLA_TIN
+    GE = DrumSound.TABLA_GE
+    DH = DrumSound.TABLA_DHA
+    TT = DrumSound.TABLA_TIT
+    KE = DrumSound.TABLA_KE
+    GB = DrumSound.TABLA_GE_BEND
+
+    # Tanpura drone — Sa + Pa
+    tanpura_sa = score.part("tanpura_sa", synth="strings_synth", envelope="pad",
+                             detune=3, lowpass=1000, volume=0.18,
+                             reverb=0.5, reverb_type="taj_mahal")
+    tanpura_pa = score.part("tanpura_pa", synth="strings_synth", envelope="pad",
+                             detune=3, lowpass=1400, volume=0.14,
+                             reverb=0.5, reverb_type="taj_mahal")
+    sa3 = Tone("Sa", octave=3, system=shruti)
+    pa3 = Tone("Pa", octave=3, system=shruti)
+    for _ in range(16):
+        tanpura_sa.add(sa3, Duration.WHOLE)
+        tanpura_pa.add(pa3, Duration.WHOLE)
+
+    # Quiet sitar — Raga Yaman (Kalyan thaat)
+    sitar = score.part("sitar", instrument="sitar", volume=0.12,
+                        reverb=0.4, reverb_type="taj_mahal")
+    ts = TonedScale(system=shruti, tonic=Tone("Sa", octave=4, system=shruti))
+    y = list(ts["kalyan"].tones)
+    S, R, G, M, P, D, N, S2 = y
+    sitar.rest(Duration.WHOLE)
+    sitar.rest(Duration.WHOLE)
+    for tone, dur, vel in [
+        (S, 3.0, 55), (R, 1.0, 50), (G, 3.0, 58), (R, 1.0, 48),
+        (S, 4.0, 55), (G, 1.0, 50), (M, 1.0, 52), (P, 3.0, 58),
+        (M, 1.0, 48), (G, 1.0, 50), (R, 1.0, 48), (S, 4.0, 55),
+        (P, 2.0, 52), (D, 1.0, 55), (N, 2.0, 58), (S2, 3.0, 60),
+        (N, 1.0, 52), (D, 1.0, 50), (P, 1.0, 52), (G, 1.0, 48),
+        (R, 1.0, 48), (S, 4.0, 55),
+    ]:
+        sitar.add(tone, dur, velocity=vel)
+
+    # 4 bars drone intro (silence for drums)
+    silence = Pattern(name="silence", time_signature="4/4", beats=16.0, hits=[])
+    score.add_pattern(silence, repeats=1)
+
+    # Gentle opening
+    p1 = Pattern(name="gentle", time_signature="4/4", beats=8.0, hits=[
+        h(DH, 0.0, 80), h(NA, 2.0, 60),
+        h(DH, 4.0, 85), h(NA, 5.0, 55), h(NA, 6.0, 60), h(DH, 7.0, 80),
+    ])
+
+    # Building with ghost notes
+    p2 = Pattern(name="build", time_signature="4/4", beats=16.0, hits=[
+        h(DH, 0.0, 95), h(TT, 0.5, 35), h(NA, 1.0, 70), h(TT, 1.5, 30),
+        h(NA, 2.0, 65), h(DH, 3.0, 90),
+        h(DH, 4.0, 100), h(TT, 4.25, 35), h(TT, 4.5, 40), h(NA, 5.0, 75),
+        h(TT, 5.5, 35), h(NA, 6.0, 70), h(TT, 6.5, 30), h(DH, 7.0, 95),
+        h(DH, 8.0, 95), h(TI, 9.0, 70), h(TI, 10.0, 72), h(NA, 11.0, 80),
+        h(TT, 11.25, 40), h(TT, 11.5, 42), h(KE, 11.75, 45),
+        h(TT, 12.0, 50), h(TT, 12.25, 55), h(KE, 12.5, 58), h(NA, 12.75, 70),
+        h(DH, 13.0, 100), h(TT, 13.25, 40), h(TT, 13.5, 45), h(KE, 13.75, 50),
+        h(NA, 14.0, 75), h(KE, 14.25, 50), h(DH, 14.5, 85), h(NA, 14.75, 70),
+        h(DH, 15.0, 110), h(GB, 15.5, 100),
+    ])
+
+    # Full intensity
+    p3 = Pattern(name="fire", time_signature="4/4", beats=16.0, hits=[
+        h(TT, 0.0, 50), h(TT, 0.125, 35), h(TT, 0.25, 45), h(KE, 0.5, 55),
+        h(NA, 0.75, 85),
+        h(DH, 1.0, 115), h(TT, 1.25, 38), h(DH, 1.5, 70), h(NA, 1.75, 60),
+        h(TT, 2.0, 50), h(TT, 2.125, 35), h(TT, 2.25, 48), h(KE, 2.5, 55),
+        h(NA, 2.75, 88),
+        h(DH, 3.0, 115), h(GB, 3.5, 105), h(NA, 3.75, 72),
+        h(NA, 4.0, 115), h(NA, 4.25, 60), h(TT, 4.5, 40), h(NA, 4.75, 105),
+        h(GE, 5.0, 105), h(GE, 5.25, 55), h(GB, 5.5, 95), h(GE, 5.75, 50),
+        h(NA, 6.0, 115), h(TT, 6.125, 30), h(TT, 6.25, 38), h(NA, 6.5, 100),
+        h(TT, 6.625, 32), h(TT, 6.75, 42),
+        h(GB, 7.0, 115), h(KE, 7.25, 52), h(GE, 7.5, 72), h(KE, 7.75, 48),
+        # Tihai
+        h(DH, 8.0, 115), h(NA, 8.25, 78), h(TT, 8.5, 52), h(KE, 8.75, 58),
+        h(DH, 9.0, 105),
+        h(DH, 9.5, 110), h(NA, 9.75, 78), h(TT, 10.0, 52), h(KE, 10.25, 58),
+        h(DH, 10.5, 105),
+        h(DH, 11.0, 120), h(NA, 11.25, 82), h(TT, 11.5, 58), h(KE, 11.75, 62),
+        h(DH, 12.0, 120),
+        # Silence... then finish
+        h(GB, 14.5, 120),
+        h(DH, 15.5, 127), h(DH, 15.75, 127),
+    ])
+
+    score.add_pattern(p1, repeats=1)
+    score.add_pattern(p2, repeats=1)
+    score.add_pattern(p3, repeats=1)
+    score.set_drum_effects(reverb=0.4, reverb_type="taj_mahal")
+
+    play_song(score, "Tabla Solo — Raga Yaman (22-Shruti, Taj Mahal)")
+
+
+def journey():
+    """Journey — piano → orchestra → world → sitar EDM.
+
+    One reverb space (Taj Mahal), tanpura drone throughout. Piano opens
+    alone, cello joins, harp/oboe/flute take over with djembe, sitar
+    arrives over tabla, builds to an EDM section with house drums.
+    """
+    REV = "taj_mahal"
+    score = Score("4/4", bpm=72)
+
+    # ── Drone — runs the entire piece ──
+    tanpura = score.part("tanpura", synth="strings_synth", envelope="pad",
+                          detune=3, lowpass=1000, volume=0.12,
+                          reverb=0.5, reverb_type=REV)
+    for _ in range(40):
+        tanpura.add("A2", Duration.WHOLE)
+
+    # ── Bars 1-8: Piano alone, then cello ──
+    piano = score.part("piano", instrument="piano", volume=0.35,
+                        reverb=0.35, reverb_type=REV)
+    for notes in [
+        ["A2","E3","A3","C4","E4","C4","A3","E3"],
+        ["F2","C3","F3","A3","C4","A3","F3","C3"],
+        ["G2","D3","G3","B3","D4","B3","G3","D3"],
+        ["E2","B2","E3","G#3","B3","G#3","E3","B2"],
+        ["A2","E3","A3","C4","E4","C4","A3","E3"],
+        ["D2","A2","D3","F3","A3","F3","D3","A2"],
+        ["E2","B2","E3","G#3","B3","G#3","E3","B2"],
+        ["A2","E3","A3","C4","E4","A4","E4","C4"],
+    ]:
+        for n in notes:
+            piano.add(n, Duration.EIGHTH, velocity=68)
+
+    cello = score.part("cello", instrument="cello", volume=0.2,
+                        reverb=0.4, reverb_type=REV)
+    cello.rest(Duration.WHOLE)
+    for note, dur, vel in [
+        ("A3", 4.0, 55), ("C4", 4.0, 58),
+        ("B3", 2.0, 52), ("A3", 2.0, 50), ("G3", 4.0, 55),
+        ("F3", 4.0, 52), ("E3", 4.0, 55), ("A3", 4.0, 58),
+    ]:
+        cello.add(note, dur, velocity=vel)
+
+    # ── Bars 9-16: Harp + oboe + flute + djembe ──
+    harp = score.part("harp", instrument="harp", volume=0.28,
+                       reverb=0.45, reverb_type=REV)
+    oboe = score.part("oboe", instrument="oboe", volume=0.22,
+                       reverb=0.4, reverb_type=REV)
+    flute = score.part("flute", instrument="flute", volume=0.18,
+                        reverb=0.4, reverb_type=REV)
+    for _ in range(8):
+        harp.rest(Duration.WHOLE)
+    for notes in [
+        ["A3","C4","E4","A4","C5","E5","A5","E5"],
+        ["D3","F3","A3","D4","F4","A4","D5","A4"],
+        ["E3","G3","B3","E4","G4","B4","E5","B4"],
+        ["A3","C4","E4","A4","E5","C5","A4","E4"],
+        ["F3","A3","C4","F4","A4","C5","F5","C5"],
+        ["E3","G#3","B3","E4","G#4","B4","E5","B4"],
+    ]:
+        for n in notes:
+            harp.add(n, Duration.EIGHTH, velocity=58)
+    for _ in range(9):
+        oboe.rest(Duration.WHOLE)
+    for note, dur, vel in [
+        ("E5", 1.5, 62), ("D5", 0.5, 55), ("C5", 1.0, 58),
+        ("A4", 1.0, 62), ("G4", 1.0, 55), ("A4", 1.5, 58),
+        ("B4", 0.5, 52), ("C5", 2.0, 62), ("A4", 4.0, 58),
+    ]:
+        oboe.add(note, dur, velocity=vel)
+    for _ in range(10):
+        flute.rest(Duration.WHOLE)
+    for note, dur, vel in [
+        ("A5", 2.0, 50), ("G5", 1.0, 45), ("E5", 1.0, 48),
+        ("D5", 2.0, 50), ("C5", 1.0, 45), ("A4", 1.0, 48),
+        ("G4", 2.0, 50), ("A4", 2.0, 52),
+    ]:
+        flute.add(note, dur, velocity=vel)
+
+    # ── Bars 15-20: Sitar + tabla ──
+    sitar = score.part("sitar", instrument="sitar", volume=0.2,
+                        reverb=0.35, reverb_type=REV)
+    for _ in range(14):
+        sitar.rest(Duration.WHOLE)
+    for note, dur, vel in [
+        ("A4", 1.0, 70), ("Bb4", 0.5, 60), ("A4", 0.5, 65),
+        ("C5", 1.5, 75), ("Bb4", 0.5, 60),
+        ("D5", 1.0, 70), ("E5", 1.5, 78),
+        ("F5", 0.5, 62), ("E5", 1.0, 70),
+        ("D5", 0.5, 62), ("C5", 0.5, 65), ("Bb4", 0.5, 58),
+        ("A4", 2.0, 75),
+        ("Bb4", 0.25, 55), ("C5", 0.25, 58), ("D5", 0.25, 62),
+        ("E5", 0.25, 68),
+        ("F5", 0.25, 65), ("G5", 0.25, 70), ("A5", 0.5, 80),
+        ("G5", 0.25, 62), ("F5", 0.25, 58), ("E5", 0.5, 62),
+        ("C5", 0.5, 58), ("Bb4", 0.5, 55), ("A4", 2.0, 75),
+    ]:
+        sitar.add(note, dur, velocity=vel)
+
+    # ── EDM section — sitar over house beat ──
+    # Solo sections: 8+8+8+12 = 36 beats = 9 bars
+    # Total bars before EDM: 8 piano + 6 harp + 6 djembe + 4 tabla + 9 solo = 33
+    edm_start = 33
+    pad = score.part("pad", instrument="synth_pad", volume=0.18,
+                      reverb=0.45, reverb_type=REV,
+                      sidechain=0.6, sidechain_release=0.15)
+    for _ in range(edm_start):
+        pad.rest(Duration.WHOLE)
+    for sym in ["Am", "F", "G", "Em"] * 2:
+        pad.add(Chord.from_symbol(sym), Duration.WHOLE)
+
+    sub = score.part("sub", instrument="808_bass", volume=0.4)
+    for _ in range(edm_start):
+        sub.rest(Duration.WHOLE)
+    for n in ["A1","A1","F1","F1","G1","G1","E1","E1"] * 2:
+        sub.add(n, Duration.HALF)
+
+    sitar2 = score.part("sitar2", instrument="sitar", volume=0.4,
+                         reverb=0.3, reverb_type=REV)
+    for _ in range(edm_start + 2):
+        sitar2.rest(Duration.WHOLE)
+    for note, dur, vel in [
+        ("A4", 0.25, 75), ("C5", 0.25, 78), ("E5", 0.5, 85),
+        ("D5", 0.25, 72), ("C5", 0.25, 70), ("A4", 0.5, 75),
+        ("G4", 0.25, 68), ("A4", 0.25, 72), ("C5", 0.5, 78),
+        ("A4", 0.5, 72),
+        ("E5", 0.5, 82), ("D5", 0.25, 72), ("C5", 0.25, 70),
+        ("A4", 0.5, 75), ("G4", 0.5, 68), ("A4", 1.0, 78),
+    ] * 2:
+        sitar2.add(note, dur, velocity=vel)
+
+    # Drums: djembe bars 9-14, tabla bars 15-20, house bars 21-28
+    DJB = DrumSound.DJEMBE_BASS
+    DJT = DrumSound.DJEMBE_TONE
+    DJS = DrumSound.DJEMBE_SLAP
+    NA = DrumSound.TABLA_NA
+    DH = DrumSound.TABLA_DHA
+    TT = DrumSound.TABLA_TIT
+    GB = DrumSound.TABLA_GE_BEND
+
+    silence = Pattern(name="s", time_signature="4/4", beats=32.0, hits=[])
+    score.add_pattern(silence, repeats=1)
+    p_dj = Pattern(name="dj", time_signature="4/4", beats=8.0, hits=[
+        _Hit(DJB, 0.0, 48), _Hit(DJT, 1.0, 40), _Hit(DJT, 1.5, 35),
+        _Hit(DJS, 2.0, 45), _Hit(DJT, 3.0, 40),
+        _Hit(DJB, 4.0, 52), _Hit(DJT, 5.0, 42), _Hit(DJT, 5.5, 38),
+        _Hit(DJS, 6.0, 48), _Hit(DJT, 6.5, 35), _Hit(DJS, 7.0, 45),
+    ])
+    score.add_pattern(p_dj, repeats=3)
+    p_tab = Pattern(name="tab", time_signature="4/4", beats=8.0, hits=[
+        _Hit(DH, 0.0, 80), _Hit(TT, 0.5, 30), _Hit(NA, 1.0, 65),
+        _Hit(NA, 2.0, 60), _Hit(DH, 3.0, 80),
+        _Hit(DH, 4.0, 85), _Hit(TT, 4.25, 32), _Hit(TT, 4.5, 35),
+        _Hit(NA, 5.0, 68), _Hit(TT, 5.5, 30), _Hit(NA, 6.0, 65),
+        _Hit(DH, 7.0, 85),
+    ])
+    score.add_pattern(p_tab, repeats=2)
+
+    # Tabla solo — everything drops out, just tabla and drone
+    KE = DrumSound.TABLA_KE
+    TI = DrumSound.TABLA_TIN
+    GE = DrumSound.TABLA_GE
+    T3 = 1.0 / 12.0   # 32nd triplet
+    T9 = 1.0 / 9.0    # ninth note
+
+    # Part 1: whisper — space, breath, single hits
+    p_solo1 = Pattern(name="solo1", time_signature="4/4", beats=8.0, hits=[
+        _Hit(DH, 0.0, 78),
+        _Hit(NA, 2.0, 55),
+        _Hit(DH, 4.0, 82),
+        _Hit(TT, 5.0, 30), _Hit(NA, 5.5, 52), _Hit(TT, 6.0, 28),
+        _Hit(DH, 7.0, 78),
+    ])
+    score.add_pattern(p_solo1, repeats=1)
+
+    # Part 2: ghosts emerge — 16th note ghost fills between accents
+    p_solo2 = Pattern(name="solo2", time_signature="4/4", beats=8.0, hits=[
+        _Hit(DH, 0.0, 92), _Hit(TT, 0.25, 32), _Hit(TT, 0.5, 35),
+        _Hit(NA, 1.0, 68), _Hit(TT, 1.25, 30), _Hit(TT, 1.5, 28),
+        _Hit(NA, 2.0, 62), _Hit(TT, 2.5, 32), _Hit(DH, 3.0, 88),
+        _Hit(TT, 3.25, 35), _Hit(TT, 3.5, 38),
+        _Hit(DH, 4.0, 95), _Hit(TT, 4.25, 38), _Hit(TT, 4.5, 42),
+        _Hit(NA, 5.0, 72), _Hit(TT, 5.25, 32), _Hit(TT, 5.5, 35),
+        _Hit(KE, 5.75, 40),
+        _Hit(NA, 6.0, 68), _Hit(TT, 6.25, 35), _Hit(KE, 6.5, 38),
+        _Hit(DH, 7.0, 95), _Hit(GB, 7.5, 88),
+    ])
+    score.add_pattern(p_solo2, repeats=1)
+
+    # Part 3: call and response — dayan vs bayan, dynamics wide open
+    p_solo3 = Pattern(name="solo3", time_signature="4/4", beats=8.0, hits=[
+        # Dayan speaks
+        _Hit(NA, 0.0, 110), _Hit(NA, 0.25, 55), _Hit(TT, 0.5, 38),
+        _Hit(NA, 0.75, 100),
+        # Bayan answers
+        _Hit(GE, 1.0, 100), _Hit(GE, 1.25, 50), _Hit(GB, 1.5, 90),
+        _Hit(GE, 1.75, 45),
+        # Dayan louder
+        _Hit(NA, 2.0, 115), _Hit(TT, 2.125, 30), _Hit(TT, 2.25, 35),
+        _Hit(NA, 2.5, 105), _Hit(TT, 2.625, 32), _Hit(TT, 2.75, 38),
+        # Bayan louder
+        _Hit(GB, 3.0, 112), _Hit(KE, 3.25, 50), _Hit(GE, 3.5, 68),
+        _Hit(KE, 3.75, 45),
+        # Together — explosion
+        _Hit(DH, 4.0, 120), _Hit(TT, 4.25, 45), _Hit(TT, 4.5, 48),
+        _Hit(DH, 5.0, 115), _Hit(TT, 5.25, 42), _Hit(NA, 5.5, 72),
+        # 9-tuplet — the weird one, breaks the grid
+        *[_Hit(TT if i % 2 == 0 else KE, 6.0 + i * T9, 38 + i * 5)
+          for i in range(9)],
+        _Hit(DH, 7.0, 118),
+    ])
+    score.add_pattern(p_solo3, repeats=1)
+
+    # Part 4: blazing — 32nd triplets, cascades, tihai finale
+    p_solo4 = Pattern(name="solo4", time_signature="4/4", beats=12.0, hits=[
+        # 32nd triplet opening flourish
+        *[_Hit(TT, 0.0 + i * T3, 40 + i * 2) for i in range(12)],
+        _Hit(DH, 1.0, 120), _Hit(GB, 1.5, 108),
+        # Rapid alternating hands
+        _Hit(NA, 2.0, 110), _Hit(KE, 2.125, 45), _Hit(NA, 2.25, 105),
+        _Hit(KE, 2.375, 48), _Hit(NA, 2.5, 108), _Hit(KE, 2.625, 50),
+        _Hit(NA, 2.75, 112),
+        _Hit(DH, 3.0, 118),
+        # Another 32nd triplet burst — longer, crescendo
+        *[_Hit(TT, 3.5 + i * T3, 32 + i * 4) for i in range(18)],
+        _Hit(DH, 5.0, 122), _Hit(DH, 5.25, 118), _Hit(GB, 5.5, 115),
+        # 9 against 4 polyrhythm moment
+        _Hit(GE, 6.0, 88), _Hit(GE, 7.0, 85),
+        *[_Hit(NA if i % 3 == 0 else TT, 6.0 + i * (2.0 / 9.0), 42 + (i % 3) * 15)
+          for i in range(9)],
+        # Grand tihai — 3x pattern, each louder
+        _Hit(DH, 8.0, 108), _Hit(NA, 8.25, 72), _Hit(TT, 8.5, 48),
+        _Hit(KE, 8.75, 52), _Hit(DH, 9.0, 100),
+        _Hit(DH, 9.25, 112), _Hit(NA, 9.5, 78), _Hit(TT, 9.75, 52),
+        _Hit(KE, 10.0, 58), _Hit(DH, 10.25, 108),
+        _Hit(DH, 10.5, 120), _Hit(NA, 10.75, 85), _Hit(TT, 11.0, 58),
+        _Hit(KE, 11.25, 62), _Hit(DH, 11.5, 127),
+        # Silence.....
+        # SLAM
+        _Hit(GB, 11.875, 127),
+    ])
+    score.add_pattern(p_solo4, repeats=1)
+
+    score.drums("house", repeats=8)
+    score.set_drum_effects(reverb=0.3, reverb_type=REV)
+
+    play_song(score, "Journey — Piano → World → Sitar EDM (Taj Mahal)")
+
+
 SONGS = {
     "1": ("Bossa Nova in A minor", bossa_nova_girl),
     "2": ("Bebop in Bb major", bebop_in_bb),
@@ -1258,6 +1574,8 @@ SONGS = {
     "20": ("Temple Bell (Japanese)", temple_bell),
     "21": ("Cinematic Showcase (Orchestral)", cinematic_showcase),
     "22": ("Greensleeves (Renaissance Lute)", greensleeves),
+    "23": ("Tabla Solo (Raga Yaman)", tabla_solo_yaman),
+    "24": ("Journey (Western → World → Indian)", journey),
 }
 
 if __name__ == "__main__":
@@ -1271,7 +1589,7 @@ if __name__ == "__main__":
             print(f"    {key:>2}. {name}")
 
         print()
-        choice = input("  Pick a song (1-22, or 'all'): ").strip()
+        choice = input("  Pick a song (1-24, or 'all'): ").strip()
         print()
 
         if choice == "all":

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "pytheory"
-version = "0.34.0"
+version = "0.35.1"
 description = "Music Theory for Humans"
 readme = "README.md"
 license = "MIT"
@@ -21,7 +21,6 @@ classifiers = [
     "Programming Language :: Python :: 3.13",
 ]
 dependencies = [
-    "pytuning",
     "numeral",
     "sounddevice",
     "scipy",

pytheory/__init__.py

@@ -1,6 +1,6 @@
 """PyTheory: Music Theory for Humans."""
 
-__version__ = "0.34.0"
+__version__ = "0.35.1"
 
 from .tones import Tone, Interval
 from .systems import System, SYSTEMS, TET

pytheory/_statics.py

@@ -1,4 +1,4 @@
-from pytuning import scales
+import math
 
 REFERENCE_A = 440
 
@@ -6,41 +6,59 @@ REFERENCE_A = 440
 # Scientific pitch notation changes octave at C, not A, so this offset
 # is needed for all octave arithmetic.
 C_INDEX = 3
-def _create_just_intonation_scale(n):
-    """5-limit just intonation ratios for 12-tone systems.
 
-    These are the pure frequency ratios derived from the harmonic series —
-    the way intervals "want" to sound before equal temperament imposed
-    compromise. Each ratio is mathematically exact: a perfect fifth is
-    exactly 3/2, a major third is exactly 5/4.
 
-    For non-12 systems, falls back to equal temperament.
+# ── Temperament scale generators (replaces pytuning dependency) ──────────
+
+def _create_edo_scale(n):
+    """N-tone equal division of the octave. Each step = 2^(1/n)."""
+    return [2 ** (i / n) for i in range(n + 1)]
+
+
+def _create_pythagorean_scale(n):
+    """Pythagorean tuning — spiral of pure fifths (3/2 ratio).
+
+    Each tone is generated by stacking perfect fifths and octave-reducing.
     """
-    from fractions import Fraction
+    ratios = [1.0]
+    for i in range(1, n):
+        # Stack fifths: (3/2)^i, then reduce to within one octave
+        r = (3 / 2) ** i
+        while r >= 2.0:
+            r /= 2.0
+        ratios.append(r)
+    ratios.sort()
+    ratios.append(2.0)
+    return ratios
+
+
+def _create_quarter_comma_meantone_scale(n):
+    """Quarter-comma meantone — pure major thirds (5/4), tempered fifths.
+
+    The fifth is narrowed by 1/4 of a syntonic comma so that four
+    fifths make a pure major third (5/4). The meantone fifth =
+    5^(1/4) ≈ 1.49535.
+    """
+    fifth = 5 ** 0.25  # meantone fifth ≈ 1.49535 (vs 1.5 pure)
+    ratios = [1.0]
+    for i in range(1, n):
+        r = fifth ** i
+        while r >= 2.0:
+            r /= 2.0
+        ratios.append(r)
+    ratios.sort()
+    ratios.append(2.0)
+    return ratios
+def _create_just_intonation_scale(n):
+    """5-limit just intonation ratios for 12-tone systems."""
     if n != 12:
-        return scales.create_edo_scale(n)
-    # Standard 5-limit JI ratios (A-based: A=1/1)
-    ratios = [
-        Fraction(1, 1),       # A  — unison
-        Fraction(16, 15),     # A# — minor second
-        Fraction(9, 8),       # B  — major second
-        Fraction(6, 5),       # C  — minor third
-        Fraction(5, 4),       # C# — major third
-        Fraction(4, 3),       # D  — perfect fourth
-        Fraction(45, 32),     # D# — augmented fourth
-        Fraction(3, 2),       # E  — perfect fifth
-        Fraction(8, 5),       # F  — minor sixth
-        Fraction(5, 3),       # F# — major sixth
-        Fraction(9, 5),       # G  — minor seventh
-        Fraction(15, 8),      # G# — major seventh
-        Fraction(2, 1),       # A  — octave
-    ]
-    return [float(r) for r in ratios]
+        return _create_edo_scale(n)
+    return [1, 16/15, 9/8, 6/5, 5/4, 4/3, 45/32, 3/2, 8/5, 5/3, 9/5, 15/8, 2.0]
 
 TEMPERAMENTS = {
-    "equal": scales.create_edo_scale,
-    "pythagorean": scales.create_pythagorean_scale,
-    "meantone": scales.create_quarter_comma_meantone_scale,
+    "equal": _create_edo_scale,
+    "pythagorean": _create_pythagorean_scale,
+    "meantone": _create_quarter_comma_meantone_scale,
     "just": _create_just_intonation_scale,
 }
 
@@ -295,8 +313,51 @@ DEGREES_SHRUTI = [
     ("shadja", ()),                 # Sa (octave)
 ]
 
+# 22-shruti frequency ratios — 5-limit just intonation.
+# These are the REAL shruti intervals, NOT 22-TET approximations.
+# Based on the traditional Pythagorean/harmonic ratios from Indian
+# musicological treatises (Natya Shastra, Sangita Ratnakara).
+#
+# Ordered from Dha (A=1.0) to match our system indexing.
+# Sa is at index 5 (ratio ≈ 6/5 from Dha).
+from fractions import Fraction
+_SHRUTI_RATIOS_FROM_SA = [
+    Fraction(1, 1),       #  0: Sa        — 1/1
+    Fraction(256, 243),   #  1: atikomal Re — Pythagorean limma
+    Fraction(16, 15),     #  2: komal Re   — JI minor second
+    Fraction(10, 9),      #  3: shuddha Re — minor whole tone
+    Fraction(9, 8),       #  4: Re         — major whole tone
+    Fraction(32, 27),     #  5: atikomal Ga — Pythagorean minor 3rd
+    Fraction(6, 5),       #  6: komal Ga   — JI minor 3rd
+    Fraction(5, 4),       #  7: Ga         — JI major 3rd
+    Fraction(81, 64),     #  8: tivra Ga   — Pythagorean major 3rd
+    Fraction(4, 3),       #  9: Ma         — perfect 4th
+    Fraction(27, 20),     # 10: ekashruti Ma
+    Fraction(45, 32),     # 11: tivra Ma   — augmented 4th
+    Fraction(729, 512),   # 12: atitivra Ma — Pythagorean tritone
+    Fraction(3, 2),       # 13: Pa         — perfect 5th
+    Fraction(128, 81),    # 14: atikomal Dha — Pythagorean minor 6th
+    Fraction(8, 5),       # 15: komal Dha  — JI minor 6th
+    Fraction(5, 3),       # 16: shuddha Dha
+    Fraction(27, 16),     # 17: Dha        — Pythagorean major 6th
+    Fraction(16, 9),      # 18: komal Ni   — Pythagorean minor 7th
+    Fraction(9, 5),       # 19: shuddha Ni — JI minor 7th
+    Fraction(15, 8),      # 20: Ni         — JI major 7th
+    Fraction(243, 128),   # 21: tivra Ni   — Pythagorean major 7th
+]
+
+# Rotate to start from Dha (index 17 in the Sa-based list above).
+# Dha = 27/16 from Sa. We divide all ratios by 27/16 and wrap.
+_dha_ratio = _SHRUTI_RATIOS_FROM_SA[17]
+SHRUTI_RATIOS = []
+for i in range(22):
+    sa_idx = (i + 17) % 22  # rotate: Dha=0, komalNi=1, ..., Sa=5, ...
+    r = _SHRUTI_RATIOS_FROM_SA[sa_idx] / _dha_ratio
+    if r < 1:
+        r *= 2  # wrap into the same octave
+    SHRUTI_RATIOS.append(float(r))
+
 # 22-shruti thaat scales with proper microtonal intervals.
-# Each interval is counted in shrutis (22-TET steps).
 # Compare to the 12-TET approximations in INDIAN_SCALES which lose
 # the distinction between 2-shruti and 3-shruti steps.
 SHRUTI_SCALES = {
@@ -341,13 +402,75 @@ SHRUTI_SCALES = {
     ],
 }
 
-# ── 24-TET Arabic maqam system ─────────────────────────────────────────────
-# Arabic maqam uses quarter-tones (half-flat, half-sharp). 24-TET captures
-# these intervals exactly. Each step = 50 cents (vs 100 in 12-TET).
-# The half-flat (♭½) is the defining sound of Arabic music — it's what
-# makes maqam Rast and Bayati sound distinctly Middle Eastern.
+# ── Arabic maqam system ───────────────────────────────────────────────────
+# Arabic maqam uses quarter-tones with specific JI ratios, NOT equal
+# 24-TET divisions. The neutral intervals (quarter-flat, quarter-sharp)
+# are based on ratios involving the 11th partial, as theorized by
+# Zalzal (8th century Baghdad). The quarter-flat E in Rast is 27/22,
+# not simply halfway between Eb and E.
 #
+# 24 positions per octave, but with unequal JI spacing.
 # Ordered from La (=A) to match Western index positions.
+
+# Maqam JI ratios from Do (C). Based on traditional practice:
+# - Standard JI intervals for the 12 chromatic positions
+# - Zalzalian ratios (11-limit) for the quarter-tone positions
+_MAQAM_RATIOS_FROM_DO = [
+    Fraction(1, 1),       #  0: Do       — unison
+    Fraction(33, 32),     #  1: Do↑      — quarter-sharp (~53¢, 33rd harmonic)
+    Fraction(16, 15),     #  2: Reb      — JI minor 2nd
+    Fraction(12, 11),     #  3: Re↓      — Zalzalian neutral 2nd (~151¢)
+    Fraction(9, 8),       #  4: Re       — major whole tone
+    Fraction(11, 9) * Fraction(1, 1),  #  5: Re↑  — undecimal (~347¢... too high)
+    Fraction(6, 5),       #  6: Mib      — JI minor 3rd
+    Fraction(27, 22),     #  7: Mi↓      — Zalzalian neutral 3rd (~355¢) THE Rast note
+    Fraction(5, 4),       #  8: Mi       — JI major 3rd
+    Fraction(4, 3),       #  9: Fa       — perfect 4th
+    Fraction(11, 8),      # 10: Fa↑      — undecimal tritone (~551¢)
+    Fraction(45, 32),     # 11: Fa#      — augmented 4th
+    Fraction(22, 15),     # 12: Sol↓     — neutral (~663¢... adjusted)
+    Fraction(3, 2),       # 13: Sol      — perfect 5th
+    Fraction(99, 64),     # 14: Sol↑     — quarter-sharp 5th
+    Fraction(8, 5),       # 15: Lab      — JI minor 6th
+    Fraction(18, 11),     # 16: La↓      — Zalzalian neutral 6th
+    Fraction(5, 3),       # 17: La       — JI major 6th
+    Fraction(27, 16),     # 18: La↑/Sib↓ — Pythagorean major 6th
+    Fraction(16, 9),      # 19: Sib      — Pythagorean minor 7th
+    Fraction(11, 6),      # 20: Si↓      — undecimal neutral 7th
+    Fraction(15, 8),      # 21: Si       — JI major 7th
+    Fraction(243, 128),   # 22: Si↑      — Pythagorean major 7th
+    Fraction(2, 1) * Fraction(33, 64),  # 23: near-octave (~1049¢)
+]
+
+# Ratios directly from La (A=1/1), each position defined explicitly.
+# Standard JI intervals for chromatic positions, Zalzalian (11-limit)
+# ratios for the quarter-tone positions.
+MAQAM_RATIOS = [
+    1.0,                              #  0: La       — A (unison)
+    float(Fraction(256, 243)),        #  1: La↑      — Pythagorean comma up
+    float(Fraction(16, 15)),          #  2: Sib      — Bb (JI minor 2nd)
+    float(Fraction(12, 11)),          #  3: Si↓      — B quarter-flat (Zalzalian)
+    float(Fraction(9, 8)),            #  4: Si       — B (major 2nd)
+    float(Fraction(6, 5)),            #  5: Do       — C (minor 3rd from A)
+    float(Fraction(11, 9)),           #  6: Do↑      — C quarter-sharp (undecimal)
+    float(Fraction(5, 4)),            #  7: Reb      — Db (major 3rd from A...= JI Db)
+    float(Fraction(9, 7)),            #  8: Re↓      — D quarter-flat (septimal)
+    float(Fraction(4, 3)),            #  9: Re       — D (perfect 4th from A)
+    float(Fraction(11, 8)),           # 10: Re↑      — D quarter-sharp (undecimal)
+    float(Fraction(45, 32)),          # 11: Mib      — Eb (augmented 4th from A)
+    float(Fraction(6, 5) * Fraction(27, 22)),  # 12: Mi↓ — E quarter-flat (Do × 27/22)
+    float(Fraction(3, 2)),            # 13: Mi       — E (perfect 5th from A)
+    float(Fraction(8, 5)),            # 14: Fa       — F (minor 6th from A)
+    float(Fraction(18, 11)),          # 15: Fa↑      — F quarter-sharp (Zalzalian)
+    float(Fraction(5, 3)),            # 16: Fa#      — F# (major 6th from A)
+    float(Fraction(27, 16)),          # 17: Sol↓     — G quarter-flat
+    float(Fraction(16, 9)),           # 18: Sol      — G (minor 7th from A)
+    float(Fraction(11, 6)),           # 19: Sol↑     — G quarter-sharp (undecimal)
+    float(Fraction(15, 8)),           # 20: Lab      — Ab (major 7th from A)
+    float(Fraction(27, 14)),          # 21: La↓      — A quarter-flat (septimal)
+    float(Fraction(243, 128)),        # 22: La½b     — near-octave
+    float(Fraction(2, 1) * Fraction(256, 257)),  # 23: La♮  — near-octave
+]
 TONES_ARABIC_24 = [
     ("La",),                #  0 — A
     ("La↑",),               #  1 — A quarter-sharp

pytheory/play.py

@@ -2,11 +2,24 @@ from enum import Enum
 import time
 
 import numpy
-import scipy.signal
 
 from .tones import Tone
 
 
+class _LazyModule:
+    """Lazy import wrapper — module loaded on first attribute access."""
+    def __init__(self, name):
+        self._name = name
+        self._mod = None
+    def __getattr__(self, attr):
+        if self._mod is None:
+            import importlib
+            self._mod = importlib.import_module(self._name)
+        return getattr(self._mod, attr)
+
+scipy = type('scipy', (), {'signal': _LazyModule('scipy.signal')})()
+
+
 def _get_sd():
     """Lazy import sounddevice — only needed for actual audio playback."""
     import sounddevice as sd
@@ -257,7 +270,7 @@ def strings_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
 
     # Delayed vibrato: ramps in over ~200ms, like a real bow
     vib_rate = 5.2 + rng.uniform(-0.3, 0.3)  # slight randomness per note
-    vib_depth = hz * 0.003  # ~5 cents
+    vib_depth = hz * 0.001  # subtle
     vib_onset = numpy.clip(t / 0.2, 0.0, 1.0)  # ramp over 200ms
     vibrato = vib_depth * vib_onset * numpy.sin(2 * numpy.pi * vib_rate * t)
 
@@ -442,7 +455,7 @@ def flute_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
 
     # Vibrato — develops after ~200ms
     vib_onset = numpy.clip(t / 0.2, 0.0, 1.0)
-    vib = hz * 0.003 * vib_onset * numpy.sin(2 * numpy.pi * 5.0 * t)
+    vib = hz * 0.0008 * vib_onset * numpy.sin(2 * numpy.pi * 5.0 * t)
 
     # Tube resonance — mostly fundamental + weak odd harmonics
     wave = numpy.sin(2 * numpy.pi * (hz + vib) * t) * 0.7
@@ -481,7 +494,7 @@ def trumpet_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
 
     # Vibrato
     vib_onset = numpy.clip(t / 0.15, 0.0, 1.0)
-    vib = hz * 0.004 * vib_onset * numpy.sin(2 * numpy.pi * 5.5 * t)
+    vib = hz * 0.001 * vib_onset * numpy.sin(2 * numpy.pi * 5.5 * t)
 
     # Lip buzz — additive with brass spectral shape
     # Trumpet has strong even AND odd harmonics (unlike clarinet)
@@ -523,7 +536,7 @@ def clarinet_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
     rng = numpy.random.default_rng(int(hz * 100) % 2**31)
 
     vib_onset = numpy.clip(t / 0.3, 0.0, 1.0)
-    vib = hz * 0.002 * vib_onset * numpy.sin(2 * numpy.pi * 4.5 * t)
+    vib = hz * 0.001 * vib_onset * numpy.sin(2 * numpy.pi * 4.5 * t)
 
     # Cylindrical bore: odd harmonics dominate
     wave = numpy.zeros(n_samples, dtype=numpy.float64)
@@ -592,7 +605,7 @@ def oboe_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
     rng = numpy.random.default_rng(int(hz * 100) % 2**31)
 
     vib_onset = numpy.clip(t / 0.2, 0.0, 1.0)
-    vib = hz * 0.004 * vib_onset * numpy.sin(2 * numpy.pi * 5.0 * t)
+    vib = hz * 0.001 * vib_onset * numpy.sin(2 * numpy.pi * 5.0 * t)
 
     wave = numpy.zeros(n_samples, dtype=numpy.float64)
     n_harmonics = min(18, int((SAMPLE_RATE / 2) / hz))
@@ -668,7 +681,7 @@ def cello_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
 
     # Delayed vibrato
     vib_rate = 5.0 + rng.uniform(-0.3, 0.3)
-    vib_depth = hz * 0.004
+    vib_depth = hz * 0.001
     vib_onset = numpy.clip(t / 0.25, 0.0, 1.0)
     vibrato = vib_depth * vib_onset * numpy.sin(2 * numpy.pi * vib_rate * t)
 
@@ -787,6 +800,336 @@ def upright_bass_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
     return (peak * out).astype(numpy.int16)
 
 
+def timpani_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
+    """Timpani — large kettle drum with definite pitch.
+
+    The copper kettle creates a tuned resonance with inharmonic
+    overtones. The head modes are at ratios 1.0, 1.5, 1.99, 2.44
+    (not integer multiples like strings). The felt mallet gives a
+    soft attack with a deep, booming body.
+    """
+    t = numpy.arange(n_samples, dtype=numpy.float64) / SAMPLE_RATE
+
+    # Timpani head modes — inharmonic but definite pitch
+    # Mode ratios from vibrating circular membrane physics
+    wave = numpy.sin(2 * numpy.pi * hz * t) * 0.8
+    wave += numpy.sin(2 * numpy.pi * hz * 1.5 * t) * 0.35 * numpy.exp(-6 * t)
+    wave += numpy.sin(2 * numpy.pi * hz * 1.99 * t) * 0.2 * numpy.exp(-10 * t)
+    wave += numpy.sin(2 * numpy.pi * hz * 2.44 * t) * 0.1 * numpy.exp(-15 * t)
+
+    # Two-stage decay: initial thump fades fast, fundamental rings
+    decay = numpy.where(t < 0.15,
+                        numpy.exp(-4 * t),
+                        numpy.exp(-4 * 0.15) * numpy.exp(-1.5 * (t - 0.15)))
+    wave *= decay
+
+    # Felt mallet impact — warm, not sharp
+    mallet_len = min(int(SAMPLE_RATE * 0.02), n_samples)
+    rng = numpy.random.default_rng(int(hz * 100) % 2**31)
+    mallet = rng.uniform(-0.3, 0.3, mallet_len)
+    mallet *= numpy.exp(-numpy.linspace(0, 8, mallet_len))
+    wave[:mallet_len] += mallet
+
+    # Copper kettle resonance — boosts low-mids
+    import scipy.signal as _sig
+    lo, hi = max(20, int(hz * 0.7)), min(SAMPLE_RATE // 2 - 1, int(hz * 2))
+    if lo < hi:
+        bp, ap = _sig.butter(2, [lo, hi], btype='band', fs=SAMPLE_RATE)
+        kettle = _sig.lfilter(bp, ap, wave) * 0.3
+        wave += kettle
+
+    mx = numpy.abs(wave).max()
+    if mx > 0:
+        wave /= mx
+
+    return (peak * wave).astype(numpy.int16)
+
+
+def saxophone_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
+    """Saxophone — single reed through a conical brass bore.
+
+    The conical bore produces all harmonics (like oboe), but the
+    brass body and larger mouthpiece give a warmer, fatter, more
+    vocal quality. The reed adds a slight buzz. Saxophone is
+    between clarinet (odd harmonics) and oboe (nasal even+odd) —
+    it has everything, with a strong fundamental and rich mids.
+    """
+    t = numpy.arange(n_samples, dtype=numpy.float64) / SAMPLE_RATE
+    rng = numpy.random.default_rng(int(hz * 100) % 2**31)
+
+    # Vibrato — develops after ~250ms, wider than flute
+    vib_onset = numpy.clip(t / 0.25, 0.0, 1.0)
+    vib = hz * 0.0012 * vib_onset * numpy.sin(2 * numpy.pi * 5.2 * t)
+
+    wave = numpy.zeros(n_samples, dtype=numpy.float64)
+    n_harmonics = min(20, int((SAMPLE_RATE / 2) / hz))
+
+    for n in range(1, n_harmonics + 1):
+        f_n = hz * n
+        if f_n >= SAMPLE_RATE / 2:
+            break
+        # Sax spectral shape: strong fundamental, broad mid peak (3-6),
+        # slower rolloff than oboe (brass body carries harmonics further)
+        if n == 1:
+            amp = 1.0
+        elif n <= 3:
+            amp = 0.6
+        elif n <= 6:
+            amp = 0.4 * numpy.exp(-0.1 * (n - 4) ** 2)
+        else:
+            amp = 0.2 / n
+        phase = rng.uniform(0, 2 * numpy.pi)
+        wave += amp * numpy.sin(2 * numpy.pi * (f_n + vib * n) * t + phase)
+
+    # Reed buzz — more present than oboe but still warm
+    reed = rng.normal(0, 0.07, n_samples)
+    # Bandpass the reed noise around 1-3kHz (the "honk" range)
+    import scipy.signal as _sig
+    reed_lo = max(20, int(hz * 2))
+    reed_hi = min(SAMPLE_RATE // 2 - 1, int(hz * 6))
+    if reed_lo < reed_hi:
+        br, ar = _sig.butter(2, [reed_lo, reed_hi], btype='band', fs=SAMPLE_RATE)
+        reed = _sig.lfilter(br, ar, reed).astype(numpy.float64) * 2.0
+    wave += reed
+
+    # Brass body warmth — low-mid boost
+    center = min(1500, hz * 4)
+    bw = 500
+    lo = max(20, int(center - bw))
+    hi = min(SAMPLE_RATE // 2 - 1, int(center + bw))
+    if lo < hi:
+        bp, ap = _sig.butter(2, [lo, hi], btype='band', fs=SAMPLE_RATE)
+        body = _sig.lfilter(bp, ap, wave) * 0.2
+        wave += body
+
+    mx = numpy.abs(wave).max()
+    if mx > 0:
+        wave /= mx
+
+    return (peak * wave).astype(numpy.int16)
+
+
+def vocal_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE, lyric="ah"):
+    """Vocal/formant synthesis — sings vowel sounds at a given pitch.
+
+    Models the human voice with:
+    1. LF glottal model — asymmetric pulse with sharp closure (not just sines)
+    2. 5 parallel resonant formant filters (real voice has 5 formant peaks)
+    3. Jitter + shimmer (natural pitch/amplitude irregularity)
+    4. Aspiration noise mixed with the glottal source
+    5. Consonant onsets (plosives, sibilants, nasals, etc.)
+    """
+    import scipy.signal as _sig
+
+    # 5-formant table: (F1, F2, F3, F4, F5) frequencies and bandwidths
+    # Based on Peterson & Barney (1952) measurements, male voice
+    FORMANTS = {
+        'a': [(800, 130), (1200, 100), (2500, 140), (3300, 250), (3750, 300)],
+        'e': [(530, 80),  (1850, 100), (2500, 130), (3300, 250), (3750, 300)],
+        'i': [(280, 60),  (2250, 100), (2900, 120), (3350, 250), (3750, 300)],
+        'o': [(500, 100), (1000, 80),  (2500, 140), (3300, 250), (3750, 300)],
+        'u': ((325, 70),  (700, 60),   (2530, 140), (3300, 250), (3750, 300)),
+    }
+    # Formant gains (relative amplitude per formant)
+    FGAINS = [1.0, 0.8, 0.5, 0.25, 0.15]
+
+    rng = numpy.random.default_rng(int(hz * 100 + len(lyric) * 7) % 2**31)
+    t = numpy.arange(n_samples, dtype=numpy.float64) / SAMPLE_RATE
+
+    # Parse vowels from lyric
+    vowels_in_lyric = [c.lower() for c in lyric if c.lower() in FORMANTS]
+    if not vowels_in_lyric:
+        vowels_in_lyric = ['a']
+
+    # ── Glottal source: LF model approximation ──
+    # Asymmetric pulse: slow open phase, sharp closure, then closed phase.
+    # Much more "voice-like" than a sine or sawtooth.
+    # Jitter (pitch irregularity) + shimmer (amplitude irregularity)
+    jitter = rng.normal(0, hz * 0.001, n_samples)  # ~0.1% pitch jitter
+    shimmer = 1.0 + rng.normal(0, 0.008, n_samples)  # ~0.8% amp shimmer
+    # Vibrato
+    vib = hz * 0.001 * numpy.sin(2 * numpy.pi * 5.5 * t)
+    inst_freq = hz + vib + jitter
+    phase = numpy.cumsum(2 * numpy.pi * inst_freq / SAMPLE_RATE)
+    # LF glottal shape: sharper falling edge via phase shaping
+    saw = (phase / (2 * numpy.pi)) % 1.0  # 0 to 1 sawtooth
+    # Asymmetric: slow rise (60%), fast fall (40%)
+    glottal = numpy.where(saw < 0.6,
+                          numpy.sin(numpy.pi * saw / 0.6),   # smooth rise
+                          -numpy.sin(numpy.pi * (saw - 0.6) / 0.4) * 0.8)  # sharp fall
+    glottal *= shimmer
+
+    # Aspiration noise (breathiness) — subtle
+    breath = rng.normal(0, 0.04, n_samples)
+    source = glottal * 0.92 + breath * 0.08
+
+    # ── Formant filtering ──
+    n_vowels = len(vowels_in_lyric)
+    out = numpy.zeros(n_samples, dtype=numpy.float64)
+
+    if n_vowels == 1:
+        # Single vowel — filter the whole thing
+        formants = FORMANTS[vowels_in_lyric[0]]
+        for (fc, bw), gain in zip(formants, FGAINS):
+            lo = max(20, fc - bw)
+            hi = min(SAMPLE_RATE // 2 - 1, fc + bw)
+            if lo < hi:
+                bp, ap = _sig.butter(2, [lo, hi], btype='band', fs=SAMPLE_RATE)
+                out += _sig.lfilter(bp, ap, source).astype(numpy.float64) * gain
+    else:
+        # Multiple vowels — crossfade formants
+        samples_per_vowel = n_samples // n_vowels
+        for vi, vowel in enumerate(vowels_in_lyric):
+            formants = FORMANTS[vowel]
+            start = vi * samples_per_vowel
+            end = n_samples if vi == n_vowels - 1 else start + samples_per_vowel
+            seg = source[start:end].copy()
+            seg_out = numpy.zeros_like(seg)
+            for (fc, bw), gain in zip(formants, FGAINS):
+                lo = max(20, fc - bw)
+                hi = min(SAMPLE_RATE // 2 - 1, fc + bw)
+                if lo < hi:
+                    bp, ap = _sig.butter(2, [lo, hi], btype='band', fs=SAMPLE_RATE)
+                    seg_out += _sig.lfilter(bp, ap, seg).astype(numpy.float64) * gain
+            # Crossfade
+            fade = min(int(SAMPLE_RATE * 0.02), len(seg_out) // 4)
+            if vi > 0 and fade > 0:
+                seg_out[:fade] *= numpy.linspace(0, 1, fade)
+            if vi < n_vowels - 1 and fade > 0:
+                seg_out[-fade:] *= numpy.linspace(1, 0, fade)
+            out[start:end] += seg_out[:end - start]
+
+    # ── Consonant onsets ──
+    lyric_lower = lyric.lower()
+    if lyric_lower and lyric_lower[0] not in 'aeiou':
+        c = lyric_lower[0]
+        cl = min(int(SAMPLE_RATE * 0.035), n_samples)
+        if c in 'tdkpb':
+            burst = rng.uniform(-0.5, 0.5, cl) * numpy.exp(-numpy.linspace(0, 18, cl))
+            out[:cl] = burst + out[:cl] * 0.2
+        elif c in 'sz':
+            sib = rng.uniform(-0.4, 0.4, cl)
+            if cl > 20:
+                bl, al = _sig.butter(2, [3000, min(8000, SAMPLE_RATE//2-1)], btype='band', fs=SAMPLE_RATE)
+                sib = _sig.lfilter(bl, al, numpy.pad(sib, (0, max(0, n_samples-cl))))[:cl]
+            sib *= numpy.exp(-numpy.linspace(0, 10, cl))
+            out[:cl] = sib * 0.6 + out[:cl] * 0.4
+        elif c in 'mn':
+            nl = min(int(SAMPLE_RATE * 0.06), n_samples)
+            nasal = numpy.sin(2*numpy.pi*250*t[:nl]) * 0.4 * numpy.exp(-numpy.linspace(0, 4, nl))
+            out[:nl] = nasal + out[:nl] * 0.4
+        elif c in 'fv':
+            fric = rng.uniform(-0.25, 0.25, cl) * numpy.exp(-numpy.linspace(0, 12, cl))
+            out[:cl] = fric * 0.5 + out[:cl] * 0.5
+        elif c in 'lr':
+            gl = min(int(SAMPLE_RATE * 0.05), n_samples)
+            ghz = hz * 0.7 + hz * 0.3 * numpy.linspace(0, 1, gl)
+            glide = numpy.sin(numpy.cumsum(2*numpy.pi*ghz/SAMPLE_RATE)) * 0.35
+            out[:gl] = glide + out[:gl] * 0.65
+        elif c == 'h':
+            hl = min(int(SAMPLE_RATE * 0.05), n_samples)
+            asp = rng.uniform(-0.4, 0.4, hl) * numpy.exp(-numpy.linspace(0, 5, hl))
+            out[:hl] = asp * 0.6 + out[:hl] * 0.4
+        elif c == 'w':
+            wl = min(int(SAMPLE_RATE * 0.06), n_samples)
+            ws = numpy.sin(numpy.cumsum(2*numpy.pi*hz/SAMPLE_RATE*numpy.ones(wl)))
+            if wl > 20:
+                bp, ap = _sig.butter(2, [max(20,300), min(800, SAMPLE_RATE//2-1)], btype='band', fs=SAMPLE_RATE)
+                ws = _sig.lfilter(bp, ap, ws)
+            ws *= numpy.linspace(0.5, 0, wl)
+            out[:wl] = ws * 0.4 + out[:wl] * 0.6
+
+    # Soft edges — prevent clicks at note boundaries
+    fade_samples = min(int(SAMPLE_RATE * 0.01), n_samples // 4)
+    if fade_samples > 0:
+        out[:fade_samples] *= numpy.linspace(0, 1, fade_samples)
+        out[-fade_samples:] *= numpy.linspace(1, 0, fade_samples)
+
+    mx = numpy.abs(out).max()
+    if mx > 0:
+        out /= mx
+
+    return (peak * out).astype(numpy.int16)
+
+
+def granular_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE,
+                  grain_size=0.04, density=50, scatter=0.5,
+                  pitch_var=12, source="saw"):
+    """Granular synthesis — clouds of tiny sound grains.
+
+    Chops a source waveform into overlapping micro-grains (10-200ms),
+    each independently windowed and optionally pitch/time scattered.
+    Creates textures impossible with other synthesis: frozen tones,
+    shimmering clouds, evolving pads, glitchy stutters.
+
+    Args:
+        hz: Base frequency.
+        grain_size: Duration of each grain in seconds (default 0.05 = 50ms).
+        density: Grains per second (default 20). Higher = denser cloud.
+        scatter: Random position jitter 0-1 (default 0.3). How much each
+            grain's read position varies from sequential order.
+        pitch_var: Random pitch variation per grain in cents (default 5).
+        source: Base waveform — ``"saw"``, ``"sine"``, ``"triangle"``,
+            ``"square"``, ``"noise"`` (default ``"saw"``).
+    """
+    rng = numpy.random.default_rng(int(hz * 100) % 2**31)
+
+    # Generate source material — longer than needed for scatter headroom
+    src_len = n_samples + int(SAMPLE_RATE * scatter * 2)
+    src_fns = {
+        "saw": sawtooth_wave, "sine": sine_wave, "triangle": triangle_wave,
+        "square": square_wave, "noise": noise_wave,
+    }
+    src_fn = src_fns.get(source, sawtooth_wave)
+    src = src_fn(hz, n_samples=src_len).astype(numpy.float64) / SAMPLE_PEAK
+
+    # Grain parameters
+    grain_samples = max(64, int(grain_size * SAMPLE_RATE))
+    n_grains = max(1, int(n_samples / SAMPLE_RATE * density))
+
+    # Hanning window for each grain (smooth fade in/out, no clicks)
+    window = numpy.hanning(grain_samples).astype(numpy.float64)
+
+    out = numpy.zeros(n_samples, dtype=numpy.float64)
+
+    for i in range(n_grains):
+        # Output position — evenly spaced with jitter
+        base_pos = int(i * n_samples / n_grains)
+        jitter = int(rng.uniform(-0.5, 0.5) * n_samples / n_grains * 0.3)
+        out_pos = max(0, min(n_samples - grain_samples, base_pos + jitter))
+
+        # Source read position — sequential with scatter
+        src_pos = int(base_pos * src_len / n_samples)
+        src_jitter = int(rng.uniform(-scatter, scatter) * grain_samples * 4)
+        src_pos = max(0, min(src_len - grain_samples, src_pos + src_jitter))
+
+        # Per-grain pitch variation via resampling
+        if pitch_var > 0:
+            cents = rng.uniform(-pitch_var, pitch_var)
+            rate = 2 ** (cents / 1200)
+            read_len = max(2, min(int(grain_samples * rate), src_len - src_pos))
+            grain_src = src[src_pos:src_pos + read_len]
+            x_old = numpy.linspace(0, 1, len(grain_src))
+            x_new = numpy.linspace(0, 1, grain_samples)
+            grain = numpy.interp(x_new, x_old, grain_src)
+        else:
+            end = min(src_pos + grain_samples, src_len)
+            grain = src[src_pos:end]
+            if len(grain) < grain_samples:
+                grain = numpy.pad(grain, (0, grain_samples - len(grain)))
+
+        # Apply window and mix
+        grain *= window[:len(grain)]
+        end = min(out_pos + len(grain), n_samples)
+        out[out_pos:end] += grain[:end - out_pos]
+
+    mx = numpy.abs(out).max()
+    if mx > 0:
+        out /= mx
+
+    return (peak * out).astype(numpy.int16)
+
+
 def acoustic_guitar_wave(hz, peak=SAMPLE_PEAK, n_samples=SAMPLE_RATE):
     """Acoustic guitar — Karplus-Strong with wooden body resonance.
 
@@ -1087,6 +1430,10 @@ class Synth(Enum):
     CELLO = "cello_synth"
     HARP = "harp_synth"
     UPRIGHT_BASS = "upright_bass_synth"
+    TIMPANI = "timpani_synth"
+    SAXOPHONE = "saxophone_synth"
+    GRANULAR = "granular_synth"
+    VOCAL = "vocal_synth"
     ACOUSTIC_GUITAR = "acoustic_guitar_synth"
     SITAR = "sitar_synth"
     ELECTRIC_GUITAR = "electric_guitar_synth"
@@ -1108,6 +1455,8 @@ _SYNTH_FUNCTIONS = {
     "marimba_synth": marimba_wave, "oboe_synth": oboe_wave,
     "harpsichord_synth": harpsichord_wave, "cello_synth": cello_wave,
     "harp_synth": harp_wave, "upright_bass_synth": upright_bass_wave,
+    "timpani_synth": timpani_wave, "saxophone_synth": saxophone_wave,
+    "granular_synth": granular_wave, "vocal_synth": vocal_wave,
     "acoustic_guitar_synth": acoustic_guitar_wave,
     "sitar_synth": sitar_wave, "electric_guitar_synth": electric_guitar_wave,
 }
@@ -1925,6 +2274,42 @@ def _synth_metal_hat(n_samples):
     return out
 
 
+def _synth_tabla_ge_bend(n_samples):
+    """Tabla Ge with upward pitch bend — palm pressing into bayan head.
+
+    The player strikes the bayan and then presses their palm into the
+    head, raising the pitch dramatically. The signature bayan sound
+    in Bollywood and fusion music.
+    """
+    t = numpy.arange(n_samples, dtype=numpy.float32) / SAMPLE_RATE
+    # Membrane thud
+    thump_len = min(int(SAMPLE_RATE * 0.07), n_samples)
+    thump_raw = _noise(thump_len)
+    if thump_len > 20:
+        bl, al = scipy.signal.butter(2, [40, 250], btype='band', fs=SAMPLE_RATE)
+        thump = scipy.signal.lfilter(bl, al, numpy.pad(thump_raw, (0, max(0, n_samples - thump_len))))[:thump_len]
+    else:
+        thump = thump_raw
+    thump *= _exp_decay(thump_len, 20) * 0.8
+    # Pitch sweep UP — 60 Hz rising to 200+ Hz as palm presses
+    # Gets quieter as pitch rises (palm mutes the head as it presses)
+    freq = 60 + 180 * (1 - numpy.exp(-4 * t))
+    phase = 2 * numpy.pi * numpy.cumsum(freq) / SAMPLE_RATE
+    body = numpy.sin(phase) * _exp_decay(n_samples, 6) * 0.9
+    # Metal shell resonance
+    metal_len = min(int(SAMPLE_RATE * 0.1), n_samples)
+    metal = numpy.sin(2 * numpy.pi * 150 * t[:metal_len]) * _exp_decay(metal_len, 8) * 0.3
+    # Sub
+    sub = _sine_f32(50, n_samples) * _exp_decay(n_samples, 5) * 0.4
+    click_len = min(250, n_samples)
+    click = _noise(click_len) * _exp_decay(click_len, 35) * 0.3
+    result = body + sub
+    result[:thump_len] += thump
+    result[:metal_len] += metal
+    result[:click_len] += click
+    return numpy.tanh(result * 1.3).astype(numpy.float32)
+
+
 def _synth_djembe_bass(n_samples):
     """Djembe bass — open palm strike in center of goatskin head.
 
@@ -2077,6 +2462,7 @@ def _render_drum_hit(sound_value, n_samples):
         DrumSound.TABLA_DHA.value: lambda n: _synth_tabla_dha(n),
         DrumSound.TABLA_TIT.value: lambda n: _synth_tabla_tit(n),
         DrumSound.TABLA_KE.value: lambda n: _synth_tabla_ke(n),
+        DrumSound.TABLA_GE_BEND.value: lambda n: _synth_tabla_ge_bend(n),
         # Dhol
         DrumSound.DHOL_DAGGA.value: lambda n: _synth_dhol_dagga(n),
         DrumSound.DHOL_TILLI.value: lambda n: _synth_dhol_tilli(n),
@@ -3246,9 +3632,54 @@ def _render_notes_to_buf(notes, buf, samples_per_beat, total_samples,
                 if analog > 0:
                     pitches = [hz * (2 ** (_rnd.gauss(0, analog * 5) / 1200))
                                for hz in pitches]
-                # Render oscillators (pass synth_kwargs for FM etc.)
-                waves = [synth_fn(hz, n_samples=n_samples, **_skw)
-                         for hz in pitches]
+                # Pitch bend: render at base pitch, then resample to shift
+                # pitch over time. Resampling preserves the synth's timbre
+                # perfectly — no sine waves, no retriggering.
+                bend_amt = getattr(note, 'bend', 0.0)
+                if bend_amt != 0:
+                    bend_type = getattr(note, 'bend_type', 'smooth')
+                    t_norm = numpy.linspace(0, 1, n_samples)
+
+                    waves = []
+                    for hz in pitches:
+                        hz_end = hz * (2 ** (bend_amt / 12))
+                        # Build pitch ratio curve (1.0 = no shift)
+                        if bend_type == 'smooth':
+                            ratio = (hz_end / hz) ** t_norm
+                        elif bend_type == 'linear':
+                            ratio = 1.0 + (hz_end / hz - 1.0) * t_norm
+                        elif bend_type == 'late':
+                            late_t = numpy.clip((t_norm - 0.6) / 0.4, 0.0, 1.0)
+                            ratio = (hz_end / hz) ** late_t
+                        else:
+                            ratio = (hz_end / hz) ** t_norm
+
+                        # Render a longer buffer at base pitch
+                        max_ratio = max(ratio.max(), 1.0)
+                        src_len = int(n_samples * max_ratio) + 100
+                        src = synth_fn(hz, n_samples=src_len, **_skw)
+                        src_f = src.astype(numpy.float64) / SAMPLE_PEAK
+
+                        # Variable-rate resampling: read through source
+                        # at speed determined by the ratio curve
+                        read_pos = numpy.cumsum(ratio)
+                        read_pos = (read_pos - read_pos[0]).astype(numpy.float64)
+                        # Clamp to source bounds
+                        read_pos = numpy.clip(read_pos, 0, src_len - 2)
+                        # Linear interpolation
+                        idx = read_pos.astype(numpy.int64)
+                        frac = read_pos - idx
+                        bent = src_f[idx] * (1 - frac) + src_f[numpy.minimum(idx + 1, src_len - 1)] * frac
+                        waves.append((bent * SAMPLE_PEAK).astype(numpy.int16))
+                else:
+                    # Per-note kwargs (e.g. lyric for vocal synth)
+                    note_skw = dict(_skw)
+                    note_lyric = getattr(note, 'lyric', '')
+                    if note_lyric:
+                        note_skw['lyric'] = note_lyric
+                    # Render oscillators
+                    waves = [synth_fn(hz, n_samples=n_samples, **note_skw)
+                             for hz in pitches]
                 # Sub-oscillator: octave-below sine
                 if sub_osc > 0:
                     for hz in pitches:
@@ -3312,6 +3743,12 @@ def _render_notes_to_buf(notes, buf, samples_per_beat, total_samples,
                     vel_cutoff = vel_to_filter * vel_scale + 1000
                     mixed = _apply_lowpass(mixed, vel_cutoff, q=filter_q)
                 end = min(start + len(mixed), total_samples)
+                # Choke: fade out any existing signal at this point
+                # so new notes don't pile up on previous tails
+                choke_len = min(int(SAMPLE_RATE * 0.003), start)
+                if choke_len > 0:
+                    fade = numpy.linspace(1.0, 0.0, choke_len).astype(numpy.float32)
+                    buf[start - choke_len:start] *= fade
                 buf[start:end] += mixed[:end - start] * volume * vel_scale
                 # Spread detuned oscillators into stereo L/R
                 if detune_up is not None and stereo_buf is not None:
@@ -3614,6 +4051,7 @@ def render_score(score):
         DrumSound.TABLA_GE.value: -0.2,
         DrumSound.TABLA_KE.value: -0.2,
         DrumSound.TABLA_DHA.value: 0.0,   # both drums = center
+        DrumSound.TABLA_GE_BEND.value: -0.2,
         # Dhol: bass left, treble right
         DrumSound.DHOL_DAGGA.value: -0.2,
         DrumSound.DHOL_TILLI.value: 0.2,
@@ -3647,6 +4085,10 @@ def render_score(score):
     drum_parts = [p for p in score.parts.values() if p.is_drums]
     for drum_part in drum_parts:
         part_stereo = numpy.zeros((total_samples, 2), dtype=numpy.float32)
+        # Track last hit position per sound for choke (new hit dampens
+        # the previous ring on the same drum)
+        _last_hit_start = {}
+
         for hit in drum_part._drum_hits:
             pos = hit.position
             if drum_swing > 0:
@@ -3663,6 +4105,21 @@ def render_score(score):
                 start = max(0, start)
             if start >= total_samples or start < 0:
                 continue
+
+            # Choke: if the same sound was hit recently, fade out
+            # the tail at this point (new hit dampens old resonance)
+            sound_id = hit.sound.value
+            if sound_id in _last_hit_start:
+                prev_start = _last_hit_start[sound_id]
+                # Quick 2ms fade-out at the new hit position
+                fade_len = min(int(SAMPLE_RATE * 0.002), max(0, start - prev_start))
+                if fade_len > 0 and start > 0:
+                    fade = numpy.linspace(1.0, 0.0, fade_len).astype(numpy.float32)
+                    fade_start = max(0, start - fade_len)
+                    for ch in range(2):
+                        part_stereo[fade_start:start, ch] *= fade
+            _last_hit_start[sound_id] = start
+
             remaining = total_samples - start
             hit_len = min(int(SAMPLE_RATE * 0.5), remaining)
             wave = _render_drum_hit(hit.sound.value, hit_len)

pytheory/rhythm.py

@@ -241,6 +241,26 @@ INSTRUMENTS = {
         "vel_to_filter": 3000,
         "analog": 0.3,
     },
+    "granular_pad": {
+        "synth": "granular_synth", "envelope": "pad",
+        "reverb": 0.4, "reverb_type": "cathedral",
+        "analog": 0.3,
+    },
+    "vocal": {
+        "synth": "vocal_synth", "envelope": "strings",
+        "reverb": 0.3, "reverb_type": "hall",
+        "humanize": 0.15,
+    },
+    "choir": {
+        "synth": "vocal_synth", "envelope": "pad",
+        "detune": 8, "spread": 0.4,
+        "reverb": 0.45, "reverb_type": "cathedral",
+    },
+    "granular_texture": {
+        "synth": "granular_synth", "envelope": "none",
+        "reverb": 0.5, "reverb_type": "taj_mahal",
+        "delay": 0.3, "delay_time": 0.4, "delay_feedback": 0.4,
+    },
     "808_bass": {
         "synth": "sine", "envelope": "pluck",
         "distortion": 0.4, "distortion_drive": 2.5,
@@ -275,6 +295,31 @@ INSTRUMENTS = {
         "fm_ratio": 2.0, "fm_index": 3.0,
         "reverb": 0.4, "reverb_type": "cathedral",
     },
+    "timpani": {
+        "synth": "timpani_synth", "envelope": "none",
+        "reverb": 0.4, "reverb_type": "cathedral",
+    },
+
+    # ── Woodwinds (continued) ──
+    "saxophone": {
+        "synth": "saxophone_synth", "envelope": "bowed",
+        "humanize": 0.15, "vel_to_filter": 1500,
+    },
+    "alto_sax": {
+        "synth": "saxophone_synth", "envelope": "bowed",
+        "humanize": 0.15, "vel_to_filter": 1800,
+    },
+    "tenor_sax": {
+        "synth": "saxophone_synth", "envelope": "bowed",
+        "lowpass": 3000,
+        "humanize": 0.15, "vel_to_filter": 1200,
+    },
+    "bari_sax": {
+        "synth": "saxophone_synth", "envelope": "bowed",
+        "lowpass": 2000,
+        "humanize": 0.15, "vel_to_filter": 800,
+        "sub_osc": 0.15,
+    },
 }
 
 
@@ -320,11 +365,19 @@ class TimeSignature:
 
 @dataclass
 class Note:
-    """A pairing of a sound (Tone, Chord, or None for rest) with a duration."""
+    """A pairing of a sound (Tone, Chord, or None for rest) with a duration.
+
+    The optional ``bend`` field specifies a pitch bend in semitones
+    applied over the note's duration. Positive = bend up, negative = down.
+    For example, ``bend=2`` bends the note up a whole step by the end.
+    """
 
     tone: object
     duration: Duration
     velocity: int = 100
+    bend: float = 0.0
+    bend_type: str = "smooth"  # "smooth" (log), "linear", "late"
+    lyric: str = ""  # syllable for vocal synth
 
     @property
     def beats(self) -> float:
@@ -421,6 +474,7 @@ class DrumSound(Enum):
     MRIDANGAM_NAM = 99   # treble ring (valanthalai/right head)
     MRIDANGAM_DIN = 100  # both heads
     MRIDANGAM_THA = 101  # muted treble
+    TABLA_GE_BEND = 108  # bayan with upward pitch bend (palm press)
     # Djembe sounds
     DJEMBE_BASS = 102    # open bass (center of head)
     DJEMBE_TONE = 103    # open tone (edge, fingers together)
@@ -2051,11 +2105,15 @@ class Part:
         self._drum_pattern_beats: float = 0.0
         self._automation: list[tuple[float, dict]] = []  # (beat, {param: value})
 
-    def add(self, tone_or_string, duration=Duration.QUARTER, *, velocity: int = 100) -> "Part":
+    def add(self, tone_or_string, duration=Duration.QUARTER, *, velocity: int = 100,
+            bend: float = 0.0, bend_type: str = "smooth", lyric: str = "") -> "Part":
         """Add a note. Accepts Tone/Chord objects or note strings like ``"E5"``.
 
         Duration can be a ``Duration`` enum or a raw float (beats).
         Velocity controls loudness (1-127, default 100).
+        Bend specifies a pitch bend in semitones over the note duration
+        (e.g. ``bend=2`` bends up a whole step, ``bend=-1`` bends down
+        a half step). Used for guitar bends, sitar meends, slides.
 
         Returns self for chaining.
         """
@@ -2064,7 +2122,9 @@ class Part:
             tone_or_string = Tone.from_string(tone_or_string, system=self._system)
         if isinstance(duration, (int, float)):
             duration = _RawDuration(duration)
-        self.notes.append(Note(tone=tone_or_string, duration=duration, velocity=velocity))
+        self.notes.append(Note(tone=tone_or_string, duration=duration,
+                               velocity=velocity, bend=bend,
+                               bend_type=bend_type, lyric=lyric))
         return self
 
     def set(self, **params) -> "Part":
@@ -2421,6 +2481,50 @@ class Part:
 
         return self
 
+    def roll(self, tone_or_string, duration=Duration.WHOLE, *,
+             velocity_start: int = 40, velocity_end: int = 100,
+             speed=Duration.SIXTEENTH) -> "Part":
+        """Play a roll — rapid repeated notes with velocity ramp.
+
+        Perfect for timpani rolls, snare rolls, tremolo on any
+        instrument. The velocity ramps from ``velocity_start`` to
+        ``velocity_end`` over the duration for crescendo/decrescendo.
+
+        Args:
+            tone_or_string: The note to repeat.
+            duration: Total duration of the roll.
+            velocity_start: Velocity of the first hit (default 40).
+            velocity_end: Velocity of the last hit (default 100).
+            speed: How fast to repeat (default SIXTEENTH notes).
+
+        Returns:
+            Self for chaining.
+
+        Example::
+
+            >>> timp = score.part("timp", instrument="timpani")
+            >>> timp.roll("C3", Duration.WHOLE, velocity_start=30, velocity_end=110)
+        """
+        if hasattr(duration, 'value'):
+            total = duration.value
+        else:
+            total = float(duration)
+        if hasattr(speed, 'value'):
+            step = speed.value
+        else:
+            step = float(speed)
+
+        n_hits = max(1, int(total / step))
+        for i in range(n_hits):
+            frac = i / max(1, n_hits - 1)
+            vel = int(velocity_start + (velocity_end - velocity_start) * frac)
+            vel = max(1, min(127, vel))
+            remaining = total - i * step
+            note_dur = min(step, remaining)
+            if note_dur > 0:
+                self.add(tone_or_string, note_dur, velocity=vel)
+        return self
+
     @property
     def is_drums(self) -> bool:
         """True if this part contains drum hits."""

pytheory/systems.py

@@ -2,8 +2,8 @@ from ._statics import (
     TEMPERAMENTS, TONES, DEGREES, SCALES,
     INDIAN_SCALES, ARABIC_SCALES, JAPANESE_SCALES,
     BLUES_SCALES, GAMELAN_SCALES, SYSTEMS,
-    TONES_SHRUTI, DEGREES_SHRUTI, SHRUTI_SCALES,
-    TONES_ARABIC_24, DEGREES_ARABIC_24, ARABIC_24_SCALES,
+    TONES_SHRUTI, DEGREES_SHRUTI, SHRUTI_SCALES, SHRUTI_RATIOS,
+    TONES_ARABIC_24, DEGREES_ARABIC_24, ARABIC_24_SCALES, MAQAM_RATIOS,
     TONES_SLENDRO, DEGREES_SLENDRO, SLENDRO_SCALES,
     TONES_PELOG, DEGREES_PELOG, PELOG_SCALES,
     TONES_THAI, DEGREES_THAI, THAI_SCALES,
@@ -14,7 +14,7 @@ from ._statics import (
 
 class System:
     def __init__(self, *, tone_names, degrees, scales=None, c_index=None,
-                 period=2.0):
+                 period=2.0, ratios=None):
         self.tone_names = tone_names
 
         self.degrees = degrees
@@ -25,6 +25,11 @@ class System:
         # 3.0 for Bohlen-Pierce (tritave).
         self.period = period
 
+        # Custom frequency ratios: if set, overrides equal temperament.
+        # A list of N floats (one per tone), each relative to the first
+        # tone (1.0). For example, just intonation shruti ratios.
+        self.ratios = ratios
+
         # c_index: the index of the "reference C" in the tone list.
         # For octave arithmetic — scientific pitch changes octave at C.
         # Default 3 for 12-TET western (A=0, A#=1, B=2, C=3).
@@ -214,6 +219,17 @@ class System:
         # descending goes in meta?
         return {"intervals": scale, "hemitonic": hemitonic, "meta": {}}
 
+    def tone(self, name, octave=4):
+        """Create a Tone in this system. Shorthand for ``Tone(name, octave=octave, system=self)``.
+
+        Example::
+
+            >>> edo19 = TET(19)
+            >>> edo19.tone(5, octave=4).frequency
+        """
+        from . import Tone
+        return Tone(name, octave=octave, system=self)
+
     def __repr__(self):
         return f"<System semitones={self.semitones!r}>"
 
@@ -352,9 +368,9 @@ SYSTEMS = {
     "31-tet": TET(31, names=_31TET_NAMES),
     # Microtonal systems with proper intervals (not 12-TET approximations)
     "shruti": System(tone_names=TONES_SHRUTI, degrees=DEGREES_SHRUTI,
-                     scales=SHRUTI_SCALES, c_index=5),
+                     scales=SHRUTI_SCALES, c_index=5, ratios=SHRUTI_RATIOS),
     "maqam": System(tone_names=TONES_ARABIC_24, degrees=DEGREES_ARABIC_24,
-                    scales=ARABIC_24_SCALES, c_index=5),
+                    scales=ARABIC_24_SCALES, c_index=5, ratios=MAQAM_RATIOS),
     "slendro": System(tone_names=TONES_SLENDRO, degrees=DEGREES_SLENDRO,
                       scales=SLENDRO_SCALES, c_index=1),
     "pelog": System(tone_names=TONES_PELOG, degrees=DEGREES_PELOG,

pytheory/tones.py

@@ -26,7 +26,7 @@ class Tone:
 
     def __init__(
         self,
-        name: str,
+        name,
         *,
         alt_names: Optional[list[str]] = None,
         octave: Optional[int] = None,
@@ -36,8 +36,10 @@ class Tone:
         """Initialize a Tone with a name, optional octave, and musical system.
 
         Args:
-            name: The note name (e.g. ``"C"``, ``"C#4"``). If the name
-                contains a digit, it is parsed as the octave.
+            name: The note name as a string (``"C"``, ``"C#4"``) or an int
+                for numbered systems (``0``, ``11``). Ints are converted to
+                strings and wrapped to the system's range (e.g. 22 in a
+                22-tone system becomes 0 at octave+1).
             alt_names: Alternate spellings for this tone (e.g. enharmonics).
             octave: The octave number. Overrides any octave parsed from *name*.
             system: The tuning system, either as a string key (``"western"``)
@@ -46,6 +48,23 @@ class Tone:
         if alt_names is None:
             alt_names = []
 
+        # Int tone names: wrap to system range, adjust octave
+        if isinstance(name, int):
+            if isinstance(system, str):
+                from .systems import SYSTEMS
+                _sys = SYSTEMS[system]
+            else:
+                _sys = system
+            n_tones = len(_sys.tone_names)
+            if name < 0 or name >= n_tones:
+                extra_octaves = name // n_tones
+                name = name % n_tones
+                if octave is None:
+                    octave = 4 + extra_octaves
+                else:
+                    octave += extra_octaves
+            name = str(name)
+
         if isinstance(name, str):
             # Normalize unicode music symbols to ASCII equivalents
             name = (name
@@ -70,6 +89,35 @@ class Tone:
                     if octave is None:
                         octave = parsed_octave
 
+        # Octave boundary fix: B#→C should increment octave,
+        # Cb→B should decrement octave (scientific pitch changes at C).
+        # Only applies to Western-style systems with letter names.
+        if octave is not None and name and name[0].isalpha():
+            if isinstance(system, str):
+                from .systems import SYSTEMS
+                _sys_check = SYSTEMS.get(system)
+            else:
+                _sys_check = system
+            if _sys_check is not None:
+                resolved = _sys_check.resolve_name(name)
+                if resolved is not None and resolved != name:
+                    orig_letter = name[0].upper()
+                    res_letter = resolved[0].upper()
+                    # Sharp crossing B→C: B# resolves to C, octave up
+                    if orig_letter == 'B' and res_letter == 'C' and '#' in name:
+                        octave += 1
+                    # Double sharp: A## resolves to B — no boundary cross
+                    # But B## resolves to C# — boundary cross
+                    if orig_letter == 'B' and res_letter not in ('B', 'A') and '##' in name:
+                        octave += 1
+                    # Flat crossing C→B: Cb resolves to B, octave down
+                    if orig_letter == 'C' and res_letter == 'B' and 'b' in name and name != 'C':
+                        octave -= 1
+                    # Double flat: D♭♭ resolves to C — no boundary cross
+                    # But C♭♭ resolves to Bb — boundary cross
+                    if orig_letter == 'C' and res_letter not in ('C', 'D') and 'bb' in name:
+                        octave -= 1
+
         self.name = name
         self.octave = octave
         self.alt_names = alt_names
@@ -762,11 +810,13 @@ class Tone:
         period = getattr(self.system, 'period', 2.0)
         c_idx = getattr(self.system, 'c_index', C_INDEX)
 
-        if period != 2.0 and temperament == "equal":
-            # Non-octave period (e.g. Bohlen-Pierce tritave=3.0):
-            # generate ratios as period^(n/tones) instead of 2^(n/tones)
-            import sympy
-            pitch_scale = [period ** sympy.Rational(i, tones) for i in range(tones + 1)]
+        # Custom ratios override temperament (e.g. shruti just ratios)
+        custom_ratios = getattr(self.system, 'ratios', None)
+        if custom_ratios is not None:
+            pitch_scale = list(custom_ratios) + [period]
+        elif period != 2.0 and temperament == "equal":
+            # Non-octave period (e.g. Bohlen-Pierce tritave=3.0)
+            pitch_scale = [period ** (i / tones) for i in range(tones + 1)]
         else:
             pitch_scale = TEMPERAMENTS[temperament](tones)
         octave = self.octave if self.octave is not None else 4
@@ -783,7 +833,7 @@ class Tone:
         if symbolic:
             return reference_pitch * ratio
         else:
-            result = reference_pitch * ratio
+            result = float(reference_pitch * ratio)
             if precision:
-                return float(result.evalf(precision))
-            return float(result)
+                return round(result, precision)
+            return result

test_pytheory.py

@@ -5320,7 +5320,7 @@ def test_supersaw_wave():
 @needs_portaudio
 def test_all_synths_in_enum():
     from pytheory.play import Synth
-    assert len(Synth) == 27
+    assert len(Synth) == 30
     for s in Synth:
         wave = s(440, n_samples=1000)
         assert len(wave) == 1000
@@ -6711,3 +6711,204 @@ def test_interval_to_non12():
     a5 = a.add(19)
     result = a.interval_to(a5)
     assert "octave" in result
+
+
+# ── Dedicated instrument synths ──────────────────────────────────────────────
+
+def test_all_dedicated_synths_render():
+    """Every dedicated synth waveform produces valid audio."""
+    from pytheory.play import (piano_wave, bass_guitar_wave, flute_wave,
+                                trumpet_wave, clarinet_wave, oboe_wave,
+                                marimba_wave, harpsichord_wave, cello_wave,
+                                harp_wave, upright_bass_wave,
+                                acoustic_guitar_wave, electric_guitar_wave,
+                                sitar_wave, SAMPLE_RATE)
+    synths = [piano_wave, bass_guitar_wave, flute_wave, trumpet_wave,
+              clarinet_wave, oboe_wave, marimba_wave, harpsichord_wave,
+              cello_wave, harp_wave, upright_bass_wave,
+              acoustic_guitar_wave, electric_guitar_wave, sitar_wave]
+    for fn in synths:
+        wave = fn(440, n_samples=11025)
+        assert len(wave) == 11025
+        assert wave.dtype == numpy.int16
+        assert numpy.abs(wave).max() > 0
+
+
+def test_piano_brightness_scales():
+    """High-pitched piano should be brighter (more high harmonics)."""
+    from pytheory.play import piano_wave
+    low = piano_wave(130, n_samples=22050)   # C3
+    high = piano_wave(1047, n_samples=22050)  # C6
+    # Both should produce valid audio
+    assert numpy.abs(low).max() > 0
+    assert numpy.abs(high).max() > 0
+
+
+def test_acoustic_guitar_body_resonance():
+    """Acoustic guitar should produce richer spectrum than raw pluck."""
+    from pytheory.play import acoustic_guitar_wave, pluck_wave
+    ag = acoustic_guitar_wave(220, n_samples=22050)
+    pk = pluck_wave(220, n_samples=22050)
+    assert len(ag) == len(pk) == 22050
+
+
+def test_cello_has_vibrato():
+    """Cello synth should produce pitch variation (vibrato)."""
+    from pytheory.play import cello_wave
+    wave = cello_wave(220, n_samples=44100)
+    assert len(wave) == 44100
+    assert numpy.abs(wave).max() > 0
+
+
+# ── Cabinet simulation ───────────────────────────────────────────────────────
+
+def test_cabinet_reduces_highs():
+    """Cabinet sim should reduce high-frequency content."""
+    from pytheory.play import _apply_cabinet
+    # White noise has flat spectrum
+    noise = numpy.random.uniform(-1, 1, 44100).astype(numpy.float32)
+    cabbed = _apply_cabinet(noise, brightness=0.5)
+    # RMS of cabbed should be lower (energy removed by filters)
+    assert numpy.sqrt(numpy.mean(cabbed ** 2)) < numpy.sqrt(numpy.mean(noise ** 2))
+
+
+def test_cabinet_brightness_param():
+    """Higher brightness = more high-frequency content passes through."""
+    from pytheory.play import _apply_cabinet
+    noise = numpy.random.uniform(-1, 1, 44100).astype(numpy.float32)
+    dark = _apply_cabinet(noise, brightness=0.0)
+    bright = _apply_cabinet(noise, brightness=1.0)
+    # Bright should have more energy than dark
+    assert numpy.sqrt(numpy.mean(bright ** 2)) > numpy.sqrt(numpy.mean(dark ** 2))
+
+
+# ── Analog drift ─────────────────────────────────────────────────────────────
+
+def test_analog_drift_varies_pitch():
+    """Analog drift should make repeated renders slightly different."""
+    from pytheory import Score, Duration
+    score1 = Score("4/4", bpm=120)
+    p1 = score1.part("t", synth="saw", analog=0.5)
+    p1.add("C4", Duration.QUARTER)
+    p1.add("C4", Duration.QUARTER)
+    # With analog > 0, each C4 gets a random pitch offset
+    # This is hard to test deterministically, just verify it renders
+    from pytheory.play import render_score
+    buf = render_score(score1)
+    assert len(buf) > 0
+
+
+# ── Guitar strumming ─────────────────────────────────────────────────────────
+
+def test_strum_requires_fretboard():
+    """Strumming without a fretboard should raise ValueError."""
+    from pytheory import Score, Duration
+    score = Score("4/4", bpm=120)
+    p = score.part("g", synth="saw")
+    with pytest.raises(ValueError, match="fretboard"):
+        p.strum("Am", Duration.QUARTER)
+
+
+def test_strum_adds_notes():
+    """Strumming should add notes to the part."""
+    from pytheory import Score, Duration, Fretboard
+    score = Score("4/4", bpm=120)
+    fb = Fretboard.guitar()
+    p = score.part("g", instrument="acoustic_guitar", fretboard=fb)
+    p.strum("Am", Duration.HALF)
+    assert len(p.notes) > 0
+
+
+def test_strum_direction():
+    """Both down and up strums should work."""
+    from pytheory import Score, Duration, Fretboard
+    score = Score("4/4", bpm=120)
+    fb = Fretboard.guitar()
+    p = score.part("g", instrument="acoustic_guitar", fretboard=fb)
+    p.strum("G", Duration.QUARTER, direction="down")
+    p.strum("G", Duration.QUARTER, direction="up")
+    assert len(p.notes) == 2
+
+
+# ── World drums ──────────────────────────────────────────────────────────────
+
+def test_tabla_sounds_render():
+    """All tabla drum sounds should produce valid audio."""
+    from pytheory.play import _render_drum_hit
+    from pytheory.rhythm import DrumSound
+    for sound in [DrumSound.TABLA_NA, DrumSound.TABLA_TIN, DrumSound.TABLA_GE,
+                  DrumSound.TABLA_DHA, DrumSound.TABLA_TIT, DrumSound.TABLA_KE]:
+        wave = _render_drum_hit(sound.value, 22050)
+        assert len(wave) == 22050
+        assert wave.dtype == numpy.float32
+
+
+def test_dhol_sounds_render():
+    from pytheory.play import _render_drum_hit
+    from pytheory.rhythm import DrumSound
+    for sound in [DrumSound.DHOL_DAGGA, DrumSound.DHOL_TILLI, DrumSound.DHOL_BOTH]:
+        wave = _render_drum_hit(sound.value, 22050)
+        assert len(wave) == 22050
+
+
+def test_mridangam_sounds_render():
+    from pytheory.play import _render_drum_hit
+    from pytheory.rhythm import DrumSound
+    for sound in [DrumSound.MRIDANGAM_THAM, DrumSound.MRIDANGAM_NAM,
+                  DrumSound.MRIDANGAM_DIN, DrumSound.MRIDANGAM_THA]:
+        wave = _render_drum_hit(sound.value, 22050)
+        assert len(wave) == 22050
+
+
+def test_djembe_sounds_render():
+    from pytheory.play import _render_drum_hit
+    from pytheory.rhythm import DrumSound
+    for sound in [DrumSound.DJEMBE_BASS, DrumSound.DJEMBE_TONE, DrumSound.DJEMBE_SLAP]:
+        wave = _render_drum_hit(sound.value, 22050)
+        assert len(wave) == 22050
+
+
+def test_metal_kit_sounds_render():
+    from pytheory.play import _render_drum_hit
+    from pytheory.rhythm import DrumSound
+    for sound in [DrumSound.METAL_KICK, DrumSound.METAL_SNARE, DrumSound.METAL_HAT]:
+        wave = _render_drum_hit(sound.value, 22050)
+        assert len(wave) == 22050
+
+
+def test_tabla_pattern_presets():
+    """All tabla patterns should load without error."""
+    from pytheory.rhythm import Pattern
+    for name in ["teental", "jhaptaal", "rupak", "dadra",
+                 "keherwa", "tabla solo", "tiri kita"]:
+        p = Pattern.preset(name)
+        assert p.beats > 0
+
+
+def test_world_drum_pattern_presets():
+    """All world drum patterns should load."""
+    from pytheory.rhythm import Pattern
+    for name in ["bhangra", "dhol chaal", "qawwali", "dholak folk",
+                 "adi talam", "mridangam korvai", "djembe", "kuku", "soli",
+                 "double kick", "metal blast", "metal groove", "metal gallop"]:
+        p = Pattern.preset(name)
+        assert p.beats > 0
+
+
+# ── Guitar presets with cabinet sim ──────────────────────────────────────────
+
+def test_guitar_presets_have_cabinet():
+    """Distorted guitar presets should have cabinet simulation."""
+    from pytheory import Score
+    for preset in ["distorted_guitar", "orange_crunch", "metal_guitar"]:
+        score = Score("4/4", bpm=120)
+        p = score.part("g", instrument=preset)
+        assert p.cabinet > 0, f"{preset} should have cabinet sim"
+
+
+def test_clean_guitar_preset():
+    from pytheory import Score
+    score = Score("4/4", bpm=120)
+    p = score.part("g", instrument="clean_guitar")
+    assert p.synth == "electric_guitar_synth"
+    assert p.cabinet > 0

uv.lock

@@ -444,15 +444,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8", size = 9979, upload-time = "2022-08-14T12:40:09.779Z" },
 ]
 
-[[package]]
-name = "mpmath"
-version = "1.3.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106, upload-time = "2023-03-07T16:47:11.061Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198, upload-time = "2023-03-07T16:47:09.197Z" },
-]
-
 [[package]]
 name = "myst-parser"
 version = "4.0.1"
@@ -707,11 +698,10 @@ wheels = [
 
 [[package]]
 name = "pytheory"
-version = "0.34.0"
+version = "0.35.1"
 source = { editable = "." }
 dependencies = [
     { name = "numeral" },
-    { name = "pytuning" },
     { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
     { name = "scipy", version = "1.17.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "sounddevice" },
@@ -732,7 +722,6 @@ docs = [
 [package.metadata]
 requires-dist = [
     { name = "numeral" },
-    { name = "pytuning" },
     { name = "scipy" },
     { name = "sounddevice" },
 ]
@@ -744,19 +733,6 @@ docs = [
     { name = "sphinx" },
 ]
 
-[[package]]
-name = "pytuning"
-version = "0.7.3"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
-    { name = "numpy", version = "2.4.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "sympy" },
-]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/26/59/e2c2fc91688f788587fb387ef6120c9a1ad3a8b88771fba9fc6a9c9a969d/PyTuning-0.7.3-py3-none-any.whl", hash = "sha256:db0b1231c012c1cf6a3c73aa7d791b4cff79a72f2ec6535f159c873fe302214b", size = 108174, upload-time = "2023-09-02T21:11:00.657Z" },
-]
-
 [[package]]
 name = "pyyaml"
 version = "6.0.3"
@@ -1151,18 +1127,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/52/a7/d2782e4e3f77c8450f727ba74a8f12756d5ba823d81b941f1b04da9d033a/sphinxcontrib_serializinghtml-2.0.0-py3-none-any.whl", hash = "sha256:6e2cb0eef194e10c27ec0023bfeb25badbbb5868244cf5bc5bdc04e4464bf331", size = 92072, upload-time = "2024-07-29T01:10:08.203Z" },
 ]
 
-[[package]]
-name = "sympy"
-version = "1.14.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "mpmath" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517", size = 7793921, upload-time = "2025-04-27T18:05:01.611Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", size = 6299353, upload-time = "2025-04-27T18:04:59.103Z" },
-]
-
 [[package]]
 name = "tomli"
 version = "2.4.0"