{"id":13418481,"url":"https://github.com/spotify/pedalboard","last_synced_at":"2026-01-16T10:03:02.381Z","repository":{"id":37596779,"uuid":"383471193","full_name":"spotify/pedalboard","owner":"spotify","description":"🎛 🔊 A Python library for audio.","archived":false,"fork":false,"pushed_at":"2026-01-12T19:21:25.000Z","size":28724,"stargazers_count":5924,"open_issues_count":107,"forks_count":313,"subscribers_count":64,"default_branch":"master","last_synced_at":"2026-01-12T21:37:05.945Z","etag":null,"topics":["audio","audio-processing","audio-production","audio-research","audio-unit","augmentation","juce","machine-learning","pybind11","python","vst3","vst3-host"],"latest_commit_sha":null,"homepage":"https://spotify.github.io/pedalboard","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/spotify.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2021-07-06T13:04:18.000Z","updated_at":"2026-01-12T19:32:13.000Z","dependencies_parsed_at":"2023-11-14T01:33:58.462Z","dependency_job_id":"e5e7b13d-498a-47e1-ab25-bd012d2202cc","html_url":"https://github.com/spotify/pedalboard","commit_stats":{"total_commits":637,"total_committers":34,"mean_commits":"18.735294117647058","dds":0.1161695447409733,"last_synced_commit":"1273304d9fdc8268d327a4ede06a1bf5ff2ac2e9"},"previous_names":[],"tags_count":80,"template":false,"template_full_name":null,"purl":"pkg:github/spotify/pedalboard","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/spotify%2Fpedalboard","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/spotify%2Fpedalboard/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/spotify%2Fpedalboard/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/spotify%2Fpedalboard/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/spotify","download_url":"https://codeload.github.com/spotify/pedalboard/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/spotify%2Fpedalboard/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28478049,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-16T06:30:42.265Z","status":"ssl_error","status_checked_at":"2026-01-16T06:30:16.248Z","response_time":107,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["audio","audio-processing","audio-production","audio-research","audio-unit","augmentation","juce","machine-learning","pybind11","python","vst3","vst3-host"],"created_at":"2024-07-30T22:01:02.766Z","updated_at":"2026-01-16T10:03:02.355Z","avatar_url":"https://github.com/spotify.png","language":"C++","readme":"![Pedalboard Logo](https://user-images.githubusercontent.com/213293/131147303-4805181a-c7d5-4afe-afb2-f591a4b8e586.png)\n\n\n[![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://github.com/spotify/pedalboard/blob/master/LICENSE)\n[![Documentation](https://img.shields.io/badge/Documentation-on%20github.io-brightgreen)](https://spotify.github.io/pedalboard)\n[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/pedalboard)](https://pypi.org/project/pedalboard)\n[![Supported Platforms](https://img.shields.io/badge/platforms-macOS%20%7C%20Windows%20%7C%20Linux-green)](https://pypi.org/project/pedalboard)\n[![Apple Silicon support for macOS and Linux (Docker)](https://img.shields.io/badge/Apple%20Silicon-macOS%20and%20Linux-brightgreen)](https://pypi.org/project/pedalboard)\n[![PyPI - Wheel](https://img.shields.io/pypi/wheel/pedalboard)](https://pypi.org/project/pedalboard)\n[![Test Badge](https://github.com/spotify/pedalboard/actions/workflows/all.yml/badge.svg)](https://github.com/spotify/pedalboard/actions/workflows/all.yml)\n[![Coverage Badge](https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/psobot/8736467e9952991ef44a67915ee7c762/raw/coverage.json)](https://gist.githubusercontent.com/psobot/8736467e9952991ef44a67915ee7c762/raw/coverage.json)\n[![PyPI - Downloads](https://img.shields.io/pypi/dm/pedalboard)](https://pypistats.org/packages/pedalboard)\n[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.7817838.svg)](https://doi.org/10.5281/zenodo.7817838)\n[![GitHub Repo stars](https://img.shields.io/github/stars/spotify/pedalboard?style=social)](https://github.com/spotify/pedalboard/stargazers)\n\n\n`pedalboard` is a Python library for working with audio: reading, writing, rendering, adding effects, and more. It supports most popular audio file formats and a number of common audio effects out of the box, and also allows the use of [VST3®](https://www.steinberg.net/en/company/technologies/vst3.html) and [Audio Unit](https://en.wikipedia.org/wiki/Audio_Units) formats for loading third-party software instruments and effects.\n\n`pedalboard` was built by [Spotify's Audio Intelligence Lab](https://research.atspotify.com/audio-intelligence/) to enable using studio-quality audio effects from within Python and TensorFlow. Internally at Spotify, `pedalboard` is used for [data augmentation](https://en.wikipedia.org/wiki/Data_augmentation) to improve machine learning models and to help power features like [Spotify's AI DJ](https://newsroom.spotify.com/2023-02-22/spotify-debuts-a-new-ai-dj-right-in-your-pocket/) and [AI Voice Translation](https://newsroom.spotify.com/2023-09-25/ai-voice-translation-pilot-lex-fridman-dax-shepard-steven-bartlett/). `pedalboard` also helps in the process of content creation, making it possible to add effects to audio without using a Digital Audio Workstation.\n\n[![Documentation](https://img.shields.io/badge/Documentation-on%20github.io-brightgreen)](https://spotify.github.io/pedalboard)\n\n## Features \n\n - Built-in audio I/O utilities ([pedalboard.io](https://spotify.github.io/pedalboard/reference/pedalboard.io.html))\n   - Support for reading and writing AIFF, FLAC, MP3, OGG, and WAV files on all platforms with no dependencies\n   - Additional support for reading AAC, AC3, WMA, and other formats depending on platform\n   - Support for on-the-fly resampling of audio files and streams with `O(1)` memory usage\n   - Live audio effects via \u003ca href=\"https://spotify.github.io/pedalboard/reference/pedalboard.io.html#pedalboard.io.AudioStream\"\u003e\u003ccode class=\"docutils literal\"\u003e\u003cspan class=\"pre\"\u003eAudioStream\u003c/span\u003e\u003c/code\u003e\u003c/a\u003e\n - Built-in support for a number of basic audio transformations, including:\n   - Guitar-style effects: `Chorus`, `Distortion`, `Phaser`, `Clipping`\n   - Loudness and dynamic range effects: `Compressor`, `Gain`, `Limiter`\n   - Equalizers and filters: `HighpassFilter`, `LadderFilter`, `LowpassFilter`\n   - Spatial effects: `Convolution`, `Delay`, `Reverb`\n   - Pitch effects: `PitchShift`\n   - Lossy compression: `GSMFullRateCompressor`, `MP3Compressor`\n   - Quality reduction: `Resample`, `Bitcrush`\n - Supports VST3® instrument and effect plugins on macOS, Windows, and Linux (\u003ca href=\"https://spotify.github.io/pedalboard/reference/pedalboard.html#pedalboard.load_plugin\"\u003e\u003ccode class=\"docutils literal\"\u003e\u003cspan class=\"pre\"\u003epedalboard.load_plugin\u003c/span\u003e\u003c/code\u003e\u003c/a\u003e)\n - Supports instrument and effect Audio Units on macOS\n - Strong thread-safety, memory usage, and speed guarantees\n   - Releases Python's Global Interpreter Lock (GIL) to allow use of multiple CPU cores\n     - No need to use `multiprocessing`!\n   - Even when only using one thread:\n     - Processes audio up to **300x** faster than [pySoX](https://github.com/rabitt/pysox) for single transforms, and 2-5x faster than [SoxBindings](https://github.com/pseeth/soxbindings) (via [iCorv](https://github.com/iCorv/pedalboard_with_tfdata))\n     - Reads audio files up to **4x** faster than [librosa.load](https://librosa.org/doc/main/generated/librosa.load.html) (in many cases)\n - Tested compatibility with TensorFlow - can be used in `tf.data` pipelines!\n\n## Installation\n\n`pedalboard` is available via PyPI (via [Platform Wheels](https://packaging.python.org/guides/distributing-packages-using-setuptools/#platform-wheels)):\n```\npip install pedalboard\n```\n\nIf you are new to Python, follow [INSTALLATION.md](https://github.com/spotify/pedalboard/blob/master/INSTALLATION.md) for a robust guide.\n\n### Compatibility\n\n`pedalboard` is thoroughly tested with Python 3.10, 3.11, 3.12, 3.13, and 3.14.\n\n- Linux\n  - Tested heavily in production use cases at Spotify\n  - Tested automatically on GitHub with VSTs\n  - Platform `manylinux` and `musllinux` wheels built for `x86_64` (Intel/AMD) and `aarch64` (ARM/Apple Silicon)\n  - Most Linux VSTs require a relatively modern Linux installation (with glibc \u003e 2.27)\n- macOS\n  - Tested manually with VSTs and Audio Units\n  - Tested automatically on GitHub with VSTs\n  - Platform wheels available for both Intel and Apple Silicon\n  - Compatible with a wide range of VSTs and Audio Units\n- Windows\n  - Tested automatically on GitHub with VSTs\n  - Platform wheels available for `amd64` (x86-64, Intel/AMD)\n\n## Examples\n\n\u003e **Note**: If you'd rather watch a video instead of reading examples or documentation, \u003cstrong\u003ewatch \u003ca href=\"https://www.youtube.com/watch?v=NYhkqXpFAlg\" target=\"_blank\"\u003e\u003cem\u003eWorking with Audio in Python (feat. Pedalboard)\u003c/em\u003e on YouTube\u003c/a\u003e\u003c/strong\u003e.\n\n### Quick start\n\n```python\nfrom pedalboard import Pedalboard, Chorus, Reverb\nfrom pedalboard.io import AudioFile\n\n# Make a Pedalboard object, containing multiple audio plugins:\nboard = Pedalboard([Chorus(), Reverb(room_size=0.25)])\n\n# Open an audio file for reading, just like a regular file:\nwith AudioFile('some-file.wav') as f:\n  \n  # Open an audio file to write to:\n  with AudioFile('output.wav', 'w', f.samplerate, f.num_channels) as o:\n  \n    # Read one second of audio at a time, until the file is empty:\n    while f.tell() \u003c f.frames:\n      chunk = f.read(f.samplerate)\n      \n      # Run the audio through our pedalboard:\n      effected = board(chunk, f.samplerate, reset=False)\n      \n      # Write the output to our output file:\n      o.write(effected)\n```\n\n\u003e **Note**: For more information about how to process audio through\n\u003e Pedalboard plugins, including how the `reset` parameter works,\n\u003e see \u003ca href=\"https://spotify.github.io/pedalboard/reference/pedalboard.html#pedalboard.Plugin.process\"\u003e\n\u003e the documentation for \u003ccode class=\"docutils literal\"\u003e\u003cspan class=\"pre\"\u003epedalboard.Plugin.process\u003c/span\u003e\u003c/code\u003e\u003c/a\u003e.\n\n### Making a guitar-style pedalboard\n\n```python\n# Don't do import *! (It just makes this example smaller)\nfrom pedalboard import *\nfrom pedalboard.io import AudioFile\n\n# Read in a whole file, resampling to our desired sample rate:\nsamplerate = 44100.0\nwith AudioFile('guitar-input.wav').resampled_to(samplerate) as f:\n  audio = f.read(f.frames)\n\n# Make a pretty interesting sounding guitar pedalboard:\nboard = Pedalboard([\n    Compressor(threshold_db=-50, ratio=25),\n    Gain(gain_db=30),\n    Chorus(),\n    LadderFilter(mode=LadderFilter.Mode.HPF12, cutoff_hz=900),\n    Phaser(),\n    Convolution(\"./guitar_amp.wav\", 1.0),\n    Reverb(room_size=0.25),\n])\n\n# Pedalboard objects behave like lists, so you can add plugins:\nboard.append(Compressor(threshold_db=-25, ratio=10))\nboard.append(Gain(gain_db=10))\nboard.append(Limiter())\n\n# ... or change parameters easily:\nboard[0].threshold_db = -40\n\n# Run the audio through this pedalboard!\neffected = board(audio, samplerate)\n\n# Write the audio back as a wav file:\nwith AudioFile('processed-output.wav', 'w', samplerate, effected.shape[0]) as f:\n  f.write(effected)\n```\n\n### Using VST3® or Audio Unit instrument and effect plugins\n\n```python\nfrom pedalboard import Pedalboard, Reverb, load_plugin\nfrom pedalboard.io import AudioFile\nfrom mido import Message # not part of Pedalboard, but convenient!\n\n# Load a VST3 or Audio Unit plugin from a known path on disk:\ninstrument = load_plugin(\"./VSTs/Magical8BitPlug2.vst3\")\neffect = load_plugin(\"./VSTs/RoughRider3.vst3\")\n\nprint(effect.parameters.keys())\n# dict_keys([\n#   'sc_hpf_hz', 'input_lvl_db', 'sensitivity_db',\n#   'ratio', 'attack_ms', 'release_ms', 'makeup_db',\n#   'mix', 'output_lvl_db', 'sc_active',\n#   'full_bandwidth', 'bypass', 'program',\n# ])\n\n# Set the \"ratio\" parameter to 15\neffect.ratio = 15\n\n# Render some audio by passing MIDI to an instrument:\nsample_rate = 44100\naudio = instrument(\n  [Message(\"note_on\", note=60), Message(\"note_off\", note=60, time=5)],\n  duration=5, # seconds\n  sample_rate=sample_rate,\n)\n\n# Apply effects to this audio:\neffected = effect(audio, sample_rate)\n\n# ...or put the effect into a chain with other plugins:\nboard = Pedalboard([effect, Reverb()])\n# ...and run that pedalboard with the same VST instance!\neffected = board(audio, sample_rate)\n```\n\n### Creating parallel effects chains\n\nThis example creates a delayed pitch-shift effect by running\nmultiple Pedalboards in parallel on the same audio. `Pedalboard`\nobjects are themselves `Plugin` objects, so you can nest them\nas much as you like:\n\n```python\nfrom pedalboard import Pedalboard, Compressor, Delay, Distortion, Gain, PitchShift, Reverb, Mix\n\npassthrough = Gain(gain_db=0)\n\ndelay_and_pitch_shift = Pedalboard([\n  Delay(delay_seconds=0.25, mix=1.0),\n  PitchShift(semitones=7),\n  Gain(gain_db=-3),\n])\n\ndelay_longer_and_more_pitch_shift = Pedalboard([\n  Delay(delay_seconds=0.5, mix=1.0),\n  PitchShift(semitones=12),\n  Gain(gain_db=-6),\n])\n\nboard = Pedalboard([\n  # Put a compressor at the front of the chain:\n  Compressor(),\n  # Run all of these pedalboards simultaneously with the Mix plugin:\n  Mix([\n    passthrough,\n    delay_and_pitch_shift,\n    delay_longer_and_more_pitch_shift,\n  ]),\n  # Add a reverb on the final mix:\n  Reverb()\n])\n```\n\n### Running Pedalboard on Live Audio\n\n`pedalboard` supports streaming live audio through\n\u003ca href=\"https://spotify.github.io/pedalboard/reference/pedalboard.io.html#pedalboard.io.AudioStream\"\u003e\nan \u003ccode class=\"docutils literal\"\u003e\u003cspan class=\"pre\"\u003eAudioStream\u003c/span\u003e\u003c/code\u003e object\u003c/a\u003e,\nallowing for real-time manipulation of audio by adding effects in Python.\n\n```python\nfrom pedalboard import Pedalboard, Chorus, Compressor, Delay, Gain, Reverb, Phaser\nfrom pedalboard.io import AudioStream\n\n# Open up an audio stream:\nwith AudioStream(\n  input_device_name=\"Apogee Jam+\",  # Guitar interface\n  output_device_name=\"MacBook Pro Speakers\"\n) as stream:\n  # Audio is now streaming through this pedalboard and out of your speakers!\n  stream.plugins = Pedalboard([\n      Compressor(threshold_db=-50, ratio=25),\n      Gain(gain_db=30),\n      Chorus(),\n      Phaser(),\n      Convolution(\"./guitar_amp.wav\", 1.0),\n      Reverb(room_size=0.25),\n  ])\n  input(\"Press enter to stop streaming...\")\n\n# The live AudioStream is now closed, and audio has stopped.\n```\n\n### Using Pedalboard in `tf.data` Pipelines\n\n```python\nimport tensorflow as tf \n\nsr = 48000 \n\n# Put whatever plugins you like in here:\nplugins = pedalboard.Pedalboard([pedalboard.Gain(), pedalboard.Reverb()]) \n\n# Make a dataset containing random noise:\n# NOTE: for real training, here's where you'd want to load your audio somehow:\nds = tf.data.Dataset.from_tensor_slices([np.random.rand(sr)])\n\n# Apply our Pedalboard instance to the tf.data Pipeline:\nds = ds.map(lambda audio: tf.numpy_function(plugins.process, [audio, sr], tf.float32)) \n\n# Create and train a (dummy) ML model on this audio:\nmodel = tf.keras.models.Sequential([tf.keras.layers.InputLayer(input_shape=(sr,)), tf.keras.layers.Dense(1)])\nmodel.compile(loss=\"mse\") \nmodel.fit(ds.map(lambda effected: (effected, 1)).batch(1), epochs=10)\n```\n\nFor more examples, see:\n - [the \"examples\" folder of this repository](https://github.com/spotify/pedalboard/tree/master/examples)\n - [the \"Pedalboard Demo\" Colab notebook](https://colab.research.google.com/drive/1bHjhJj1aCoOlXKl_lOfG99Xs3qWVrhch)\n - [_Working with Audio in Python (feat. Pedalboard)_ by Peter Sobot at EuroPython 2022](https://www.youtube.com/watch?v=NYhkqXpFAlg)\n - [an interactive web demo on Hugging Face Spaces and Gradio](https://huggingface.co/spaces/akhaliq/pedalboard) (via [@AK391](https://github.com/AK391)) \n\n## Contributing\n\nContributions to `pedalboard` are welcomed! See [CONTRIBUTING.md](https://github.com/spotify/pedalboard/blob/master/CONTRIBUTING.md) for details.\n\n## Citing\n\nTo cite `pedalboard` in academic work, use [its entry on Zenodo](https://doi.org/10.5281/zenodo.7817838): [![DOI 7817838](https://zenodo.org/badge/DOI/10.5281/zenodo.7817838.svg)](https://doi.org/10.5281/zenodo.7817838)\n\nTo cite via BibTeX:\n\n```tex\n@software{sobot_peter_2023_7817838,\n  author       = {Sobot, Peter},\n  title        = {Pedalboard},\n  month        = jul,\n  year         = 2021,\n  publisher    = {Zenodo},\n  doi          = {10.5281/zenodo.7817838},\n  url          = {https://doi.org/10.5281/zenodo.7817838}\n}\n```\n\n## License\n`pedalboard` is Copyright 2021-2025 Spotify AB.\n\n`pedalboard` is licensed under the [GNU General Public License v3](https://www.gnu.org/licenses/gpl-3.0.en.html). `pedalboard` includes a number of libraries that are statically compiled, and which carry the following licenses:\n\n - The core audio processing code is pulled from [JUCE 6](https://juce.com/), which is [dual-licensed under a commercial license and the GPLv3](https://juce.com/juce-6-licence).\n - The [VST3 SDK](https://github.com/steinbergmedia/vst3sdk), bundled with JUCE, is owned by [Steinberg® Media Technologies GmbH](https://www.steinberg.net/en/home.html) and licensed under the GPLv3.\n - The `PitchShift` plugin and `time_stretch` functions use [the Rubber Band Library](https://github.com/breakfastquay/rubberband), which is [dual-licensed under a commercial license](https://breakfastquay.com/technology/license.html) and the GPLv2 (or newer). [FFTW](https://www.fftw.org/) is also included to speed up Rubber Band, and [is licensed under the GPLv2 (or newer)](https://www.fftw.org/doc/License-and-Copyright.html).\n - The `MP3Compressor` plugin uses [libmp3lame from the LAME project](https://lame.sourceforge.io/), which is [licensed under the LGPLv2](https://github.com/lameproject/lame/blob/master/README) and [upgraded to the GPLv3 for inclusion in this project (as permitted by the LGPLv2)](https://www.gnu.org/licenses/gpl-faq.html#AllCompatibility).\n - The `GSMFullRateCompressor` plugin uses [libgsm](http://quut.com/gsm/), which is [licensed under the ISC license](https://github.com/timothytylee/libgsm/blob/master/COPYRIGHT) and [compatible with the GPLv3](https://www.gnu.org/licenses/license-list.en.html#ISC).\n - WAV files with formats not natively supported by JUCE (ADPCM, A-law, µ-law, 64-bit float) are decoded using [dr_wav](https://github.com/mackron/dr_libs) by David Reid, which is released into the [public domain](https://unlicense.org/).\n\n_VST is a registered trademark of Steinberg Media Technologies GmbH._\n","funding_links":[],"categories":["C++","Libraries","语音识别与合成_其他","Hosts \u0026 DAWs","Audio Processing \u0026 I/O"],"sub_categories":["Data Transformation and Manipulation","网络服务_其他"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fspotify%2Fpedalboard","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fspotify%2Fpedalboard","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fspotify%2Fpedalboard/lists"}