Softcatala/whisper-ctranslate2
A fast, memory-efficient command-line client for OpenAI Whisper speech-to-text model inference using CTranslate2 optimization.
Not currently ranked — collecting fresh signals.
star history
This project provides a high-performance Whisper implementation that is up to 4x faster than OpenAI’s original while using less memory, with batched inference achieving up to 16x speedup. It maintains CLI compatibility with the official OpenAI Whisper client for easier migration. The tool supports speaker diarization, voice activity detection filtering, custom fine-tuned models, and live microphone transcription.
Frequently asked
- What is Softcatala/whisper-ctranslate2?
- A fast, memory-efficient command-line client for OpenAI Whisper speech-to-text model inference using CTranslate2 optimization.
- Is whisper-ctranslate2 open source?
- Yes — Softcatala/whisper-ctranslate2 is open source, released under the MIT license.
- What language is whisper-ctranslate2 written in?
- Softcatala/whisper-ctranslate2 is primarily written in Python.
- How popular is whisper-ctranslate2?
- Softcatala/whisper-ctranslate2 has 1.3k stars on GitHub.
- Where can I find whisper-ctranslate2?
- Softcatala/whisper-ctranslate2 is on GitHub at https://github.com/Softcatala/whisper-ctranslate2.