# SYSTRAN/faster-whisper

Faster Whisper transcription with CTranslate2

Repository: https://github.com/SYSTRAN/faster-whisper
Canonical: https://ross.abutalabs.com/products/faster-whisper
Language: Python
License: MIT
License Family: permissive
Topics: deep-learning, inference, quantization, speech-recognition, speech-to-text, transformer, whisper, openai
Last push: 2025-11-19T14:40:46+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 53, release rhythm 42, longevity 92
- inputs: {"age_days": 1299, "days_push": 287, "days_rel": 306, "gap_med": 86, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 25104, forks 2045 (observed 2026-08-28T04:11:37.852682+00:00)

## What it is
A fast reimplementation of OpenAI's Whisper speech-to-text model built on the CTranslate2 inference engine, offering up to 4x speedup and lower memory usage than the original. It supports 8-bit quantization on both CPU and GPU and requires no system FFmpeg installation.

## Use cases
- transcribe audio files to text
- convert speech to text in python
- faster whisper transcription
- run whisper on cpu with low memory
- batch transcribe long audio recordings
- subtitle generation from audio
- deploy speech-to-text on gpu efficiently

## When to choose
- you need faster and more memory-efficient Whisper inference than openai/whisper
- you want int8 quantization for CPU or GPU deployment
- you want a pip-installable Python library without system FFmpeg dependencies
- you need batched transcription for throughput

## When to avoid
- you need a non-Python runtime such as C++ or a single-file binary (consider whisper.cpp)
- you need features beyond transcription like training or fine-tuning Whisper
- you need streaming real-time ASR with low-latency partial results

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, llm-inference, machine-learning, deep-learning
- domain: speech-processing, machine-learning
- platform: python, cross-platform
- tags: whisper, ctranslate2, speech-to-text, transcription, quantization, int8, asr, audio, gpu

## Member repositories
- SYSTRAN/faster-whisper (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:37.852682+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:56:06.170207+00:00, confidence not recorded.
  - readme: https://github.com/SYSTRAN/faster-whisper (fetched 2026-08-28T04:11:37.852682+00:00, sha 6b9a3142cb4d)
  - registry_pypi: https://pypi.org/pypi/faster-whisper/json (fetched 2026-08-29T07:53:03.776482+00:00, sha 1e513660eece)
- Data as of 2026-08-30T08:39:29.467469+00:00.