polyvoice

Speaker diarization for Python — who spoke when.

Rust-powered, ONNX-based speaker diarization that runs on CPU, fits in 30 MB, and requires zero Python runtime overhead. Pipeline v2 with K-means auto-k clustering and overlap detection.

Install

pip install polyvoice

Requires Python 3.9+.

Quick start

import polyvoice

# Models auto-download on first run (~30 MB)
pipeline = polyvoice.Pipeline.balanced()

result = pipeline.run(samples, sample_rate=16000)

print(f"Speakers: {result['num_speakers']}")
for turn in result["turns"]:
    print(f"Speaker {turn['speaker']}: {turn['start']:.1f}s - {turn['end']:.1f}s")

API

polyvoice.Pipeline.balanced(models_cache=None) — balanced accuracy / speed.
polyvoice.Pipeline.mobile(models_cache=None) — smaller, faster model.
pipeline.run(samples, sample_rate) → dict with num_speakers and turns.
pipeline.run_result(samples, sample_rate) → typed DiarizationResult with .to_json() / .to_rttm() / .to_srt() / .to_vtt() / .to_txt() projections.
polyvoice.DiarizationResult.from_json(json) — re-hydrate a saved result.

Performance

Pipeline	VoxConverse-test DER (collar 0, overlap-scored)	Model size
default (v2+VBx)	15.4%	~30 MB

Full protocol, collar/averaging disclosure, and competitor numbers: docs/BENCHMARKS.md.

See the full repository for Rust / C / CLI APIs, benchmarks, and development docs.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

polyvoice-0.11.0-cp312-cp312-win_amd64.whl (8.6 MB view details)

Uploaded Jul 25, 2026 CPython 3.12Windows x86-64

polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl (10.4 MB view details)

Uploaded Jul 25, 2026 CPython 3.12manylinux: glibc 2.38+ x86-64

polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl (9.3 MB view details)

Uploaded Jul 25, 2026 CPython 3.12macOS 11.0+ ARM64

File details

Details for the file polyvoice-0.11.0-cp312-cp312-win_amd64.whl.

File metadata

Download URL: polyvoice-0.11.0-cp312-cp312-win_amd64.whl
Upload date: Jul 25, 2026
Size: 8.6 MB
Tags: CPython 3.12, Windows x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for polyvoice-0.11.0-cp312-cp312-win_amd64.whl
Algorithm	Hash digest
SHA256	`715f2e12ca5b60310134b8019b5d61930019e50276ee5cdd89865195ddcca655`
MD5	`186e1d3540dcc08f63faad38d9c93162`
BLAKE2b-256	`b36b0d9e0ef48b4849094d217fe1b8314863bfa52a34b844ee280d7d3645fc59`

See more details on using hashes here.

Provenance

The following attestation bundles were made for polyvoice-0.11.0-cp312-cp312-win_amd64.whl:

Publisher: release.yml on ekhodzitsky/polyvoice

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: polyvoice-0.11.0-cp312-cp312-win_amd64.whl
- Subject digest: 715f2e12ca5b60310134b8019b5d61930019e50276ee5cdd89865195ddcca655
- Sigstore transparency entry: 2251644685
- Sigstore integration time: Jul 25, 2026
Source repository:
- Permalink: ekhodzitsky/polyvoice@f5a93f15347918256067e05cdfb7a6db83b4b015
- Branch / Tag: refs/tags/v0.11.0
- Owner: https://github.com/ekhodzitsky
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release.yml@f5a93f15347918256067e05cdfb7a6db83b4b015
- Trigger Event: push

File details

Details for the file polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl.

File metadata

Download URL: polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl
Upload date: Jul 25, 2026
Size: 10.4 MB
Tags: CPython 3.12, manylinux: glibc 2.38+ x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl
Algorithm	Hash digest
SHA256	`a61177399ed4e48b1aad56acde271af3f136e29b825ce9515a3d81832eaac71d`
MD5	`bbad025cbe0ae8e5f31ea57baea006b1`
BLAKE2b-256	`ea31779fa95320338ec89539aa1dcd76f84b3aec49d17a74bf6d766567a95066`

See more details on using hashes here.

Provenance

The following attestation bundles were made for polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl:

Publisher: release.yml on ekhodzitsky/polyvoice

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: polyvoice-0.11.0-cp312-cp312-manylinux_2_38_x86_64.whl
- Subject digest: a61177399ed4e48b1aad56acde271af3f136e29b825ce9515a3d81832eaac71d
- Sigstore transparency entry: 2251644954
- Sigstore integration time: Jul 25, 2026
Source repository:
- Permalink: ekhodzitsky/polyvoice@f5a93f15347918256067e05cdfb7a6db83b4b015
- Branch / Tag: refs/tags/v0.11.0
- Owner: https://github.com/ekhodzitsky
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release.yml@f5a93f15347918256067e05cdfb7a6db83b4b015
- Trigger Event: push

File details

Details for the file polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl.

File metadata

Download URL: polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl
Upload date: Jul 25, 2026
Size: 9.3 MB
Tags: CPython 3.12, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`964d2983b19ebf2779d1593d422aaa617fbd9b3aac4dac71928fda3a416bbf8f`
MD5	`c7edc65443af20233429a01d49a09bd3`
BLAKE2b-256	`fdf715dbe631640df9d177181e8c41cef8342aedb71a41fe57a04c187be9902a`

See more details on using hashes here.

Provenance

The following attestation bundles were made for polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl:

Publisher: release.yml on ekhodzitsky/polyvoice

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: polyvoice-0.11.0-cp312-cp312-macosx_11_0_arm64.whl
- Subject digest: 964d2983b19ebf2779d1593d422aaa617fbd9b3aac4dac71928fda3a416bbf8f
- Sigstore transparency entry: 2251644363
- Sigstore integration time: Jul 25, 2026
Source repository:
- Permalink: ekhodzitsky/polyvoice@f5a93f15347918256067e05cdfb7a6db83b4b015
- Branch / Tag: refs/tags/v0.11.0
- Owner: https://github.com/ekhodzitsky
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release.yml@f5a93f15347918256067e05cdfb7a6db83b4b015
- Trigger Event: push