whisperX and whisper-run
WhisperX is a mature, widely-adopted framework that combines Whisper ASR with word-level alignment and diarization, while whisper-run is a lightweight wrapper around Faster Whisper that adds diarization as a simpler alternative for users prioritizing speed over the comprehensive timestamp precision that WhisperX provides.
About whisperX
m-bain/whisperX
WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
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.
About whisper-run
gorkemkaramolla/whisper-run
Faster Whisper with Speaker Diarization
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work