WhisperX

by Max Bain

Fast transcription with word-level timestamps and speaker identification.

24.5 k stars GitHub BSD-2-Clause FreeOpen sourceLocal

View on GitHub →

What it does

Word-level alignment, ideal for precise subtitles
Diarization: knows who speaks when (via pyannote)
Batched inference, up to 70x real time with large-v2

Install it

Requirements: Python; CUDA toolkit 12.8 for GPU; Hugging Face token for diarization.

Install with pip

pip install whisperx

Transcribe with speakers

whisperx path/to/audio.wav --model large-v2 --diarize --highlight_words True

The prompt for your agent

Paste it into your coding agent: it checks the requirements and installs it for you.

Claude CodeCodexCursorGemini CLIopencode

Prompt

Install WhisperX (https://github.com/m-bain/whisperX) on my machine.
1. Check the requirements: Python; CUDA toolkit 12.8 for GPU; Hugging Face token for diarization. If anything is missing, tell me before installing anything.
2. Follow the official method:
- Install with pip: pip install whisperx
- Transcribe with speakers: whisperx path/to/audio.wav --model large-v2 --diarize --highlight_words True
3. Check that it works, then explain how to launch it and where to start.

Always review what your agent is about to run, especially install commands.