extract-speech
extract-speech is a Rust library for running voice activity detection (VAD) models. It accepts normalized mono 16 kHz PCM samples and returns detected speech ranges as sample offsets. An optional command-line application decodes common audio formats and writes the detected regions as clips or one concatenated file.
It supports:
- Silero VAD v5 and v6 through Candle or ONNX Runtime
- PulseVAD FP32 and INT8 graphs through Candle or ONNX Runtime
- PyAnnote segmentation models through Candle or ONNX Runtime
- FunASR FSMN-VAD FP32 and INT8 graphs through Candle or ONNX Runtime
- TEN VAD through Candle or ONNX Runtime
- NVIDIA Frame-VAD MarbleNet FP32 and INT8 graphs through Candle or ONNX Runtime
- WAV, MP3, FLAC, Ogg, Opus, M4A, and AAC input
- WAV and Ogg Opus output
- automatic stereo-to-mono conversion and sample-rate conversion
- parallel processing of a directory of audio files
- optional CUDA, TensorRT, and CoreML execution providers
- a reusable, model-independent Rust inference API
- optional Python bindings built with PyO3 and maturin
- JSON metadata containing clip durations and inference time
Library quick start
Add the library to your project:
[dependencies]
extract-speech = "0.9"
Load a Silero model through Candle and run inference on normalized mono 16 kHz samples:
use extract_speech::{
download::{AssetManager, ModelAsset},
Detector, Model, Result, Runtime, VadParams,
};
fn main() -> Result<()> {
let model = AssetManager::default_cache()?.model(ModelAsset::SileroV6)?;
let mut detector = Detector::builder(model.model_path())
.model(Model::Silero)
.runtime(Runtime::Candle)
.parameters(VadParams {
threshold: 0.7,
..VadParams::default()
})
.build()?;
let samples = vec![0.0_f32; 16_000];
for segment in detector.detect(&samples)? {
println!("speech: {}..{} samples", segment.start, segment.end);
}
Ok(())
}
The same Detector can process multiple independent inputs; model state is reset between calls. See the library guide for runtime features, ONNX Runtime initialization, and API details.
Python quick start
Use uv to build and install the PyO3 extension from the repository, then load a cached model with one call:
uv sync --no-dev
import array
import extract_speech
detector = extract_speech.Detector.from_pretrained()
samples = array.array("f", [0.0] * extract_speech.SAMPLE_RATE)
segments = detector.detect(samples)
Tagged releases publish prebuilt wheels for CPython 3.9–3.15 to PyPI and the corresponding GitHub release. See the Python guide for supported platforms, local model paths, ONNX Runtime, buffer types, and download helpers.
CLI quick start
Install Rust and Protocol Buffers first; see the installation guide for platform-specific instructions.
git clone https://github.com/RustedBytes/extract-speech.git
cd extract-speech
cargo build --release --features cli
./target/release/extract-speech download pulsevad --cache-dir .cache/extract-speech
Extract each detected speech region to a WAV file:
./target/release/extract-speech \
--model-path .cache/extract-speech/models/pulsevad-fp32/pulsevad_2.1k.onnx \
--process-audio input.wav \
--output output
Create one file with the detected regions joined together:
./target/release/extract-speech \
--model-path .cache/extract-speech/models/pulsevad-fp32/pulsevad_2.1k.onnx \
--process-audio input.wav \
--output-type concatenated \
--output speech.wav
The default model is PulseVAD, the default runtime is Candle, the default detection threshold is 0.7, and the default output sample rate is 16 kHz. Run extract-speech --help for the complete command reference.
Cargo features
| Feature | Default | Provides |
|---|---|---|
candle |
Yes | Silero, PulseVAD, PyAnnote, FSMN-VAD, TEN VAD, and MarbleNet inference through Candle |
onnxruntime |
Yes | All supported models through dynamically loaded ONNX Runtime |
download |
Yes | Checksum-verified model bundles, ONNX Runtime downloads, and persistent caching |
python |
No | PyO3 extension module with inference and download APIs |
cli |
No | The extract-speech executable and audio file I/O |
accelerate-src |
No | Apple Accelerate integration for Candle builds |
Runtimes
| Runtime | Models | External runtime library | Acceleration |
|---|---|---|---|
| Candle | Silero, PulseVAD FP32/dequantized INT8, PyAnnote, FSMN-VAD FP32/dequantized INT8, TEN VAD, MarbleNet FP32/dequantized INT8 | No | CPU |
| ONNX Runtime | Silero, PulseVAD FP32/INT8, PyAnnote, FSMN-VAD FP32/INT8, TEN VAD, MarbleNet FP32/INT8 | Yes, supplied with --dylib-path |
CPU, CUDA, TensorRT, CoreML |
For ONNX Runtime setup and compatible model requirements, see Models and runtimes.
Documentation
- Changelog — notable changes by release
- Library — Rust API, Cargo features, and inference examples
- Python API — installation, automatic model loading, and inference
- Installation — prerequisites, builds, and releases
- Usage — inputs, outputs, CLI options, metadata, and examples
- Models and runtimes — model compatibility and hardware acceleration
- Development — architecture, checks, and contribution workflow
Development
cargo fmt -- --check
cargo clippy --all-targets --all-features -- -D warnings -W clippy::pedantic
cargo test --all-targets --all-features
cargo build --features cli
uv sync --locked
uv run ruff check .
uv run ruff format --check .
uv run pyright
uv run python -m unittest discover -s tests/python -v
See the development guide for the code layout and project conventions.
Citation
@software{Smoliakov_Extract_Speech_2026,
author = {Smoliakov, Yehor},
month = sep,
title = {{extract-speech: Extract speech from audio files using Voice Activity Detection models}},
url = {https://github.com/RustedBytes/extract-speech},
version = {0.9.0},
year = {2026}
}
License
This project is available under the MIT License.
1 maintainer
Release files for extract-speech 0.9.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distributions (wheels)
Total release size: 63.1 MB
Release files / extract_speech-0.9.1-cp315-cp315-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp315-cp315-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.15 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
e9f5caca4c691cf808b0b48d31a320e44e01ce7fe3601c4357e411be5d5799fb
|
|
BLAKE2b-256 checksum How to use checksums |
b1b1fdb0b4ebdd065c15c99293bdf080c6a819b2565a1d9c126738dc5506e1d0
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp315-cp315-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp315-cp315-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.15 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
1f73b0c10a52818922098783691fb34171faf003cef947c947a7c649a965d2c1
|
|
BLAKE2b-256 checksum How to use checksums |
0782c6d800d9fc620fc25bd91623185660b1ad173f0fb662771380d410e38628
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp315-cp315-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp315-cp315-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.15 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
83027f79a226dac1d5075427bc397a7271b07bcc0543713b3f9f6ebff4cc7945
|
|
BLAKE2b-256 checksum How to use checksums |
16d266e3503e1a126af951b01cffcd33ea0d726393048b4a852320a46b138c04
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp314-cp314-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp314-cp314-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.14 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
e28b8dc4cacf7edb763ac804668030c9114b3290372079df2fc6ce7673b07697
|
|
BLAKE2b-256 checksum How to use checksums |
1d6269f4931ad5f04929e7301c111f115d8b19e8dab2bd39c5e3af77bad5a59e
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp314-cp314-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp314-cp314-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.14 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
4263e3e41723517804c86b64bb3369dd70cf04788018c13bdeab6c519f99673e
|
|
BLAKE2b-256 checksum How to use checksums |
5111e086df3302366bc7883bfe0eced717544770457291dcc3bacfa1cb5e9071
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp314-cp314-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp314-cp314-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.14 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
dd16468249196e78c01d0f6493643d9490b2ffeea5d13c72188f1d72fc739c0e
|
|
BLAKE2b-256 checksum How to use checksums |
cb96c5af7c02dbccb63fd0a4b9e33615bddf4ba9832e602983cbb5f920244800
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp313-cp313-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp313-cp313-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.13 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
a0451963cd56688ef761831832eb7463abc598761d83e117504d7315bdd649ab
|
|
BLAKE2b-256 checksum How to use checksums |
ca4c24c586df7e286e0ece112a14ed9dd418ed1e025ece64ed4c9b917077952f
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp313-cp313-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp313-cp313-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.13 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
b6ae048bbd497ee5dbfcd5b2c32f803ea52a26e48b76c730bfc3ad062b9662d2
|
|
BLAKE2b-256 checksum How to use checksums |
803c865fb65648be1591e830ad544bf6ed53da8246cec28b98529bc4c4ed998d
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp313-cp313-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp313-cp313-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.13 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
7c2fb344b26e68c537c15512bb00789774af0127506ce3719840665fec934166
|
|
BLAKE2b-256 checksum How to use checksums |
379268a8dc31070f690e833ebf87457ff3041425bd334b76172d08e8440b6921
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp312-cp312-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
33df49d83ccfdc2cf7f9cebc4f9f44bce7fa3936595aa5d7eac4e098c9d366b8
|
|
BLAKE2b-256 checksum How to use checksums |
c40d85ab3ee5edee5eebdefb3407f472fee0a18f2ef39dcab5a640db031029cd
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp312-cp312-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp312-cp312-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.12 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
b3abbeb14d4efa96ffa562f2f61fe69bec05b3b0b4deb3856f6ad79c6220c704
|
|
BLAKE2b-256 checksum How to use checksums |
0f393889e975be4249d8ae0d9159b82355ca42bd5dc41a7fdd37b8c67d1d19ae
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp312-cp312-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp312-cp312-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.12 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
bc7dcee1bbfe261a6834767ff34c4dc53b30e27b04b540879595e75344fefb66
|
|
BLAKE2b-256 checksum How to use checksums |
1253329d5bd24d5863aa868913e5206d1d8dc7bc122ca91ec6cd4238dec5fd90
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp311-cp311-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
33ef47dbd1603c58cbe48202687425be807d609c77cc5bb75b752d2522a87e8f
|
|
BLAKE2b-256 checksum How to use checksums |
71ae0b56fca93229ab0ed066bf0c30ab201b0123cf826b89807810efbf1717e2
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp311-cp311-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp311-cp311-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.11 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
5338887f4ab4c43d0da909498c29d50e89a55004e027d30967c2227d715ff16a
|
|
BLAKE2b-256 checksum How to use checksums |
c3e31c556535e9293a519ead97897edf559349a94f0beb0074679deb6bd60523
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp311-cp311-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp311-cp311-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.11 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
452ee4656973c8b419390e4e15c314bc0f6290b3fac0063aded272732c51ade0
|
|
BLAKE2b-256 checksum How to use checksums |
41a975b5e4e8b807c97b1e01b36de656057604869a41f111ffe3f1afcde9f5ab
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp310-cp310-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
57544b1af4b9006d84aa8672ea3f1140308793ee1d19e3fd01f1e84f581d8ebd
|
|
BLAKE2b-256 checksum How to use checksums |
3d08e0672cca6413ae0d231d8a31b7a88db17ffe2d238d985bffcfb1431fbf94
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp310-cp310-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp310-cp310-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.10 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
5f9decbc92adf20d042a7daba88a38332c85ded36843794c2f866d7a0d3e2187
|
|
BLAKE2b-256 checksum How to use checksums |
914404827a9dc6c67e34bfc530389d76cf6c24c87c44db9e52933378cb878af3
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp310-cp310-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp310-cp310-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.10 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
acd49c8baecf29979f902f193407c3f566f9ed84b49b479ea37f96f40df9f678
|
|
BLAKE2b-256 checksum How to use checksums |
6a32dec9c7c44cea83ab1009fa7cccf396c87a56755c4c4d50d4c0d97b1f8fd5
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp39-cp39-win_amd64.whl
| Download URL | extract_speech-0.9.1-cp39-cp39-win_amd64.whl |
|---|---|
| Size | 3.3 MB |
| Tags | CPython 3.9 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
e963c703ec505bfe136053caa0a1b8c1a2e574f75392f08905e4f054cefcabbd
|
|
BLAKE2b-256 checksum How to use checksums |
490793a1213d72c84e494819712a2d27d117bd3a05d69b47fce78bb69351fbbd
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp39-cp39-manylinux_2_28_x86_64.whl
| Download URL | extract_speech-0.9.1-cp39-cp39-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 3.2 MB |
| Tags | CPython 3.9 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
ec2412e5ed5cc2ba41389e4bd38ad5ce51e3109c365e4e78e9656e3c47abb762
|
|
BLAKE2b-256 checksum How to use checksums |
329e98e428e6df00f67b2ce047523427bcc91381be7d7b01283b57b9faab7d1f
|
| 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 Sep 23, 2026.
Transparency logRelease files / extract_speech-0.9.1-cp39-cp39-macosx_11_0_arm64.whl
| Download URL | extract_speech-0.9.1-cp39-cp39-macosx_11_0_arm64.whl |
|---|---|
| Size | 2.5 MB |
| Tags | CPython 3.9 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
155e76bd99ad9e611eed5988130c56ac1a2af8781b31becb2c94bfd5a152745a
|
|
BLAKE2b-256 checksum How to use checksums |
d9a01dcc9b87a4c76b5afc6886ad342109f089c7a01f26b0efcd7b6ab4d44f97
|
| 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 Sep 23, 2026.
Transparency log
