WhisperX Speech Recognition with Word-Level Timestamps and Diarization

WhisperX extends OpenAI Whisper with batched inference for 70x realtime transcription, phoneme-based word-level timestamp alignment via wav2vec2, voice activity detection, and speaker diarization. It produces accurate per-word timestamps and speaker labels from audio files.

WhisperX Speech Recognition with Word-Level Timestamps and Diarization

WhisperX extends OpenAI Whisper with batched inference for 70x realtime transcription, phoneme-based word-level timestamp alignment via wav2vec2, voice activity detection, and speaker diarization. It produces accurate per-word timestamps and speaker labels from audio files.

Installation

Method 1, Agent Skill Exchange

Method 2, Git clone

git clone https://github.com/agentskillexchange/skills.git && cd skills/skills/whisperx-speech-recognition-timestamps-diarization

Method 3, Download ZIP

  • Download the repository ZIP and extract skills/whisperx-speech-recognition-timestamps-diarization.

Method 4, Manual copy

  • Copy this skill folder into your local skills directory, then reload your agent tooling.

Method 5, Fork and sync

  • Fork the repository if you want to maintain local edits while syncing upstream changes.

Source