jaudiopy
jaudiopy is a complete, high‑performance library for beat making, sound synthesis, mixing, mastering, and processing existing audio files (local path or web link). It ships with a pure‑Python engine that has zero required dependencies and an optional numpy/scipy engine (user selectable) for maximum speed, plus drum machines, step sequencers, melodic synths, samplers, a vocal‑mixing chain, a one‑call full‑beat builder, and MIDI import — all through a clean, unified API.
Quick Start – Load and Process an Audio File
from jaudiopy import AudioFile
song = AudioFile("song.wav")
song.add_effect("reverb", "song_reverb.wav", room_size=0.7, mix=0.4)
song.play()
Main Capabilities
· Audio Files – Load from a local path or a web link, play in the terminal, apply effects, change speed/pitch, tune, trim, normalize, fade, mix, and master — all through the AudioFile class.
· Dual Engine – Pure‑Python engine with zero required dependencies, and a numpy engine (change_to_numpy=True) that uses scipy.signal.lfilter when available; both paths are optimized end‑to‑end and produce equivalent results.
· Drum Machine & Sequencing – Built‑in drum kits (kick, snare, hihat, and more), the X/x/. step pattern language, a StepSequencer, and a MelodySequencer for chord progressions.
· Synths & Instruments – Bass, pluck, pad, and piano synths, plus a dedicated 808 bass line generator with continuous pitch glide.
· Effects – Filters, EQ, distortion, bitcrush, delay, reverb, chorus, tremolo, vibrato, phaser, wah, saturation, exciter, noise gate, compressor, limiter, panning, autopan, and stereo widening.
· Mixing & Mastering – A Mixer/AdvancedMixer with buses, and a mastering chain with ready‑made presets (balanced, warm, bright, loud_edm, vocal_pop) including loudness targeting and stereo widening.
· Sampling – Load your own one‑shots or recordings with Sampler, re‑pitch and loop them, or chop them into slices for breaks.
· Vocals – A vocal‑ready EQ/de‑ess/compress/limit chain and automatic sidechain ducking against a beat.
· Full Beat Builder – BeatBuilder generates a complete, mixed, and mastered beat from a style, BPM, key, and length in a single call, at any duration.
· Utilities – MIDI note import, WAV I/O (mono/stereo, file or in‑memory bytes), and URL/local‑path resolution built in.
Installation
# Basic installation (core dependencies only)
pip install jaudiopy
# With the numpy engine
pip install jaudiopy[numpy]
# With numpy + scipy for maximum performance
pip install jaudiopy[fast]
# With more reliable terminal playback
pip install jaudiopy[playback]
# With mp3/ogg/... support (also requires ffmpeg)
pip install jaudiopy[formats]
# With MIDI import support
pip install jaudiopy[midi]
# All optional features
pip install jaudiopy[all]
Debug
The core library has no external dependency — it only uses array, math,
random, and wave from the Python standard library. If an optional
feature (playback, formats, MIDI) isn't working, make sure the matching
extra above is installed.
For jaudiopy[formats] (mp3/ogg/... support), you also need ffmpeg
installed on your system:
· Ubuntu/Debian: sudo apt install ffmpeg
· Termux (Android): pkg install ffmpeg
· macOS: brew install ffmpeg
Or download from ffmpeg.org
More Examples
Engine switch: numpy vs pure‑Python
from jaudiopy import AudioFile
song = AudioFile("song.wav", change_to_numpy=False) # pure-python engine (default)
song = AudioFile("song.wav", change_to_numpy=True) # numpy engine
Drums and the X/x/. Pattern
from jaudiopy import drums, StepSequencer
kit = drums.build_kit({
"K": ("kick", "808"),
"S": ("snare", "fat"),
"H": ("hihat", "trap"),
})
seq = StepSequencer(bpm=140)
track = seq.render_kit(
{"K": "X...x...X...x...", "S": "....X.......X...", "H": "X.X.X.X.X.X.X.X."},
{
"K": {"X": kit["K"], "x": (kit["K"], 0.5)},
"S": {"X": kit["S"]},
"H": {"X": kit["H"], "x": (kit["H"], 0.4)},
},
)
Building a Full Beat Fast
from jaudiopy import BeatBuilder
bb = BeatBuilder(style="trap_808", bpm=140, key="E", change_to_numpy=True)
left, right = bb.build(duration_minutes=2.5)
bb.build_and_save("my_beat.wav", duration_minutes=2.5)
Sampling
from jaudiopy import Sampler
sampler = Sampler.load("piano_note_C4.wav", base_note="C", base_octave=4)
note = sampler.play_note("D#", 4, duration=1.0) # re-pitched to D#4
loop = sampler.loop_to(8.0) # looped to 8 seconds
chops = sampler.chop(8) # 8 equal slices
Adding Vocals Over a Beat
from jaudiopy import mix_vocal_with_beat
final_left, final_right = mix_vocal_with_beat(
beat_left, beat_right, vocal_buffer, vocal_at=8.0,
)
Mastering With a Preset
from jaudiopy import master_with_preset
left, right = master_with_preset(left, right, preset="loud_edm", target_crest_db=7.0, haas_mix=0.4)
Issues and Contributions
You can report bugs via GitHub Issues or submit fixes via pull requests.
Links
· GitHub repository: https://github.com/JCode-JCode/jaudiopy
· PyPI page: https://pypi.org/project/jaudiopy/
License
This project is licensed under the Apache License 2.0 – see the LICENSE file for details.
Designed and built with love by J Code
Metadata
Release files for jaudiopy 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| jaudiopy-1.0.0.tar.gz | 46.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| jaudiopy-1.0.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 103.2 kB
Release files / jaudiopy-1.0.0.tar.gz
| Download URL | jaudiopy-1.0.0.tar.gz |
|---|---|
| Size | 46.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
05f6258871b62ec05b51c7d30e480f12273f9fbc4641e5692b267eb14fe7b059
|
|
BLAKE2b-256 checksum How to use checksums |
4a62e70b96369241996edee98f0521f2134755187b74c141abe89168ad36dbf5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 26, 2026.
Transparency logRelease files / jaudiopy-1.0.0-py3-none-any.whl
| Download URL | jaudiopy-1.0.0-py3-none-any.whl |
|---|---|
| Size | 56.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0e95cdf133db2ccb706ee39a7f19d35fddb1c3aba6f7a6c0865dbd4b55eb81bd
|
|
BLAKE2b-256 checksum How to use checksums |
d4af61e3ae0bf960dc6e902941b850ad03347b589b3611ca4666cabba63b01d8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 26, 2026.
Transparency log