m-bain/whisperX

WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)

90
/ 100
Verified

Builds on OpenAI's Whisper by combining faster-whisper for batched GPU inference (70x speedup) with wav2vec2 forced phoneme alignment to achieve sub-word timing accuracy. Integrates pyannote-audio for speaker diarization and includes VAD preprocessing to reduce hallucinations while maintaining quality. Supports multiple languages with automatic language-specific alignment model selection from HuggingFace and torchaudio.

20,758 stars and 864,629 monthly downloads. Used by 5 other packages. Actively maintained with 15 commits in the last 30 days. Available on PyPI.

Maintenance 20 / 25
Adoption 25 / 25
Maturity 25 / 25
Community 20 / 25

How are scores calculated?

Stars

20,758

Forks

2,188

Language

Python

License

BSD-2-Clause

Last pushed

Mar 17, 2026

Monthly downloads

864,629

Commits (30d)

15

Dependencies

12

Reverse dependents

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/m-bain/whisperX"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.