faster-whisper
Faster Whisper transcription with CTranslate2
faster-whisper scores 5.2/10 (Solid) in Voice & Realtime.
Language Python License MIT Stars 24k★ Type library Self-host Yes Repo SYSTRAN/faster-whisper Homepage https://github.com/SYSTRAN/faster-whisper
Quality
Solid 5.2
Adoption 60
Activity 5
Maturity 79
Community 58
Capability 96
Show the math
Overall 5.2/10
value = 5.2/10
score = value
Adoption 6.4/10
value = 59.5/100
score = 1 + 9 * value / 100
Activity 1.4/10
value = 5/100
score = 1 + 9 * value / 100
Maturity 8.1/10
value = 78.7/100
score = 1 + 9 * value / 100
Community 6.2/10
value = 58.2/100
score = 1 + 9 * value / 100
Capability 9.7/10
value = 96.3/100
score = 1 + 9 * value / 100
Key metrics
24k★ GitHub stars
0/90d Recent commits
0 devs/90d Recent contributors
3.7% WER WER, LibriSpeech test-clean (Whisper Large-V3 it reimplements) faster-whisper is a CTranslate2 reimplementation of OpenAI Whisper and produces equivalent accuracy; Whisper Large-V3 scores 3.7% WER on LibriSpeech test-clean in Picovoice's open benchmark (4.0% on CommonVoice). Reported as 100-WER. The number is inherited from the underlying Whisper model, not a faster-whisper-specific eval. Source: Picovoice speech-to-text-benchmark (https://github.com/Picovoice/speech-to-text-benchmark).
3.4y Project age
8mo ago Last commit
Gotchas
No gotchas documented yet. Contribute one if you know a constraint we missed.
Labels
- Self-hosted
Self-hosted
You deploy and operate it yourself; there is no hosted option here.
stt whisper inference quantization fast