How to use Faster Whisper: what you can do with it

3 use cases

• Add fast speech-to-text to a Python script

  Install Faster Whisper and transcribe audio files in a few lines, on GPU or on CPU with int8.

  • REPO SYSTRAN/faster-whisper
  • Claude Code
  • Codex
  • Cursor
  • Content
  • Beginner
  View prompt

  Add SYSTRAN/faster-whisper to my Python project and write a function transcribe(path) that loads the <model size> model on <cuda or cpu with int8>, transcribes the file with beam_size=5 and returns a list of (start, end, text) plus the detected language. Test it on <audio file>.

  Share Email me this

• Get word-level timestamps for subtitles and captions

  Turn on word_timestamps so every word comes back with its own start and end time.

  • REPO SYSTRAN/faster-whisper
  • Claude Code
  • Codex
  • Cursor
  • Content
  • Intermediate
  View prompt

  Using SYSTRAN/faster-whisper, transcribe <video audio file> with word_timestamps=True and the VAD filter on, then write an SRT file with one caption per short phrase and a JSON file of every word with its start and end time.

  Share Email me this

• Clear a transcription backlog with batched GPU inference

  Use BatchedInferencePipeline to transcribe long audio many times faster on one GPU.

  • REPO SYSTRAN/faster-whisper
  • Claude Code
  • Codex
  • Cursor
  • Content
  • Intermediate
  View prompt

  Write a script with SYSTRAN/faster-whisper that walks <folder> of audio files and transcribes each with BatchedInferencePipeline at batch_size 16 on the turbo model, saving a .txt per file and skipping files already done. Log the time per file.

  Share Email me this

Quick start with Faster Whisper

1. Check you have what it needs: Python 3.9 or greater; No system FFmpeg needed: audio is decoded with PyAV; For GPU: NVIDIA cuBLAS for CUDA 12 and cuDNN 9 for CUDA 12.
2. Open SYSTRAN/faster-whisper on GitHub and read the README and license.
3. Run pip install faster-whisper
4. Create a model with WhisperModel(model_size, device="cuda", compute_type="float16"), or use device cpu with compute_type int8 if you have no GPU
5. Call model.transcribe("audio.mp3", beam_size=5) to get segments and info
6. Iterate over the segments to print start, end and text; segments is a generator, so transcription only runs when you loop over it

These steps come from our top use case, Add fast speech-to-text to a Python script, which also has a copy-paste agent prompt.

Using the Deduped MCP, get the setup steps for SYSTRAN/faster-whisper

claude mcp add --transport http deduped https://dedupedai.com/mcp

Best AI repos for Content

About this repo

• Who it is forPython developers who need fast, cheap speech-to-text inside their own scripts, apps or pipelines.
• Why it is hot26k stars, talked about in 2 AI sources we track
• LanguagePython
• Sourcegithub.com/SYSTRAN/faster-whisper · Back to Hot repos

Requirements

• Python 3.9 or greater
• No system FFmpeg needed: audio is decoded with PyAV
• For GPU: NVIDIA cuBLAS for CUDA 12 and cuDNN 9 for CUDA 12

Related repos

• Trended earlier REPO ggml-org/whisper.cpp Whisper CPP whisper.cpp is a plain C/C++ port of OpenAI's Whisper speech recognition model with no dependencies. It runs fully offline on CPUs, Apple Silicon (Metal and Core ML), NVIDIA, AMD and Vulkan GPUs, supports integer quantization and voice activity detection, and ships a whisper-cli tool plus examples for iOS, Android, WebAssembly and Raspberry Pi.
• Trended earlier REPO KittenML/KittenTTS Kitten TTS Kitten-TTS is an open-source family of text-to-speech models, including the 1.7B 1-bit Kitten-TTS 2 with voice cloning and expression control and lightweight 15M to 80M models that run on CPU.
• Trended earlier REPO ZinYY/AdaFlash Ada Flash AdaFlash is an open-source framework for faster speculative decoding that pairs Qwen3 target models with on-policy distilled diffusion drafters and an adaptive length head, served through a patched SGLang.