faster-whisper
by SYSTRAN
faster-whisper is a Python library that runs OpenAI's Whisper speech-to-text model through the CTranslate2 engine. It turns audio files into timestamped text, running up to four times faster than the original and using less memory.
- Stars
- 26.2k
- Licence
- MIT
- Runs on
- Your computer
- Setup
- Moderate
- Language
- Python
- Age
- 1354 days
What you can do with it
- Transcribe audio files such as interviews, calls or podcasts into timestamped text
- Detect the spoken language of a recording automatically
- Run batched transcription on a GPU or a CPU with 8-bit quantisation to save time and memory
Make money with it
An individual could offer a low-cost transcription or subtitling service for podcasters and video creators, using this to do the bulk of the work before a quick manual check.
An idea, not financial advice. Check the licence before charging for anything built on it.
Ways a business could use it
- Transcribe meetings, client calls and training recordings in-house instead of paying per-minute transcription fees
- Build subtitle or captioning features into a video or podcast workflow to speed up delivery
Can you use it commercially?
Free to use, change and sell, including in paid products. Keep the copyright notice.
Not legal advice. Read the project's licence file before building a product on it.
Getting it running
Install it with "pip install faster-whisper" on Python 3.9 or later; GPU use also needs NVIDIA cuBLAS and cuDNN libraries for CUDA 12.
Follow the official instructions in the project README.
Things to check
- It is a code library, so you need some Python knowledge to use it; there is no ready-made interface
- GPU use depends on specific CUDA and cuDNN versions, which can be fiddly to match
- Large models are slow on a CPU without a GPU
Checked 27 Oct 2026. Summary drafted by AI from the project's README.
