#88th of 22 in Voice

whisperx

Batched Whisper transcription with word-level timestamps and speaker diarization

Stars: 24.4k
License: BSD-2-Clause
Last commit: Sep 2026
Last release: May 2026
Language: Python

Missed the latest GitHub check; these facts may be older.

Overview

WhisperX transcribes audio with Whisper through a faster-whisper backend, using VAD-based batching. It then aligns words to the audio with language-specific wav2vec2 models and labels speakers with pyannote diarization. It runs as a CLI or Python library, and the README claims 70x realtime with large-v2.

Who it is for: Engineers needing timestamped, speaker-labelled transcripts from audio

Strengths

Word-level timestamps via wav2vec2 forced alignment, not Whisper's utterance-level timing
Batched inference; README claims 70x realtime with large-v2 on GPU
large-v2 with beam_size=5 needs under 8GB GPU memory per README
Runs on CPU with --compute_type int8, including macOS

Weaknesses

Diarization needs a Hugging Face token and accepting the pyannote model agreement
Words like "2014." or "£13.60" cannot be aligned and get no timing
Overlapping speech is handled poorly; README says diarization is far from perfect
Each language needs its own wav2vec2 alignment model; unsupported ones need manual setup

What it needs

GPU optional
Needs CUDA toolkit 12.8 (GPU only), ffmpeg, Hugging Face token (diarization), pyannote-audio, faster-whisper
Models: Whisper (faster-whisper/CTranslate2), wav2vec2 alignment models, pyannote speaker-diarization-community-1, silero-vad

Also in Voice

See all 22

Also in Voice
Rank	Project	Score
1	voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models66.4k starsAGPL-3.0	82 out of 100
2	whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription54.4k starsMIT	75 out of 100
3	VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech57.5k starsMIT	70 out of 100
#4	buzzOffline desktop app for transcribing and translating audio with Whisper22k starsMIT	68 out of 100
#5	meetilyDesktop app that records, transcribes and summarizes meetings locally Live demo ↗ (opens in a new tab)31.7k starsMIT	65 out of 100

How to choose

Latency decides usability for voice agents; check the end-to-end numbers the project publishes.
Most quality models need a GPU; CPU-only setups are slower and limited to smaller models.
Check language coverage and whether models are downloaded at build or at first run.

Health

Releases: Not checked
Top contributor: Not checked
Security advisories: Not checked
Package advisories: Not checked
OpenSSF Scorecard: Not checked

Not read for this project yet: each refresh reads the facts of a few projects. Bots and coding-agent accounts are left out of the commit counts. Not part of the score: how we check.

License in plain words

BSD-2-Clause Permissive

Use it at work: Yes
Change it and rebrand it: Yes
Sell it or host it for others: Yes

Your part: Keep the copyright and license notice in copies you ship.

A plain summary, not legal advice. Read the license on GitHub ↗ (opens in a new tab)

Where to run it

Vast.ai ↗ (opens in a new tab)Rent a GPU by the hour from a marketplace of hosts.

Referral links: if you sign up through one, we get a commission. They never change a rank.

#8 of 22 in Voice on Archestack, as of 15 Oct 2026.

In your README

The badge updates twice a day.

[![Archestack](https://archestack.com/badge/m-bain/whisperx.svg)](https://archestack.com/best-of-ai/projects/whisperx/)

How badges work

On your site or blog

A card with the rank, score and signals, in light or dark to match the reader.

<iframe src="https://archestack.com/embed/best-of-ai/whisperx/" title="whisperx on Archestack" width="400" height="270" style="border:0;max-width:100%" loading="lazy"></iframe>

Preview the card

Best of Open-Source AI on GitHub

Each review comes from the project's README and facts we check on GitHub. The fix form opens with whisperx filled in.