Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
This is the project's own description from GitHub — we show it as-is rather than inventing details.
This project is listed under Speech-to-Text in Artificial Intelligence, a category made for creators, businesses, students and developers who want to use AI without getting lost in research papers.
Do I need coding knowledge? Coding Knowledge Recommended. (Estimate based on project type: A developer tool/library — setup usually follows the README with basic coding skills. Always confirm from the repo's README.)
The source code is released under the MIT license. Open-source licenses like this usually allow free use, including commercial use — but always read the actual license file in the repository before relying on it. Note: the code being free does not mean hosted services, APIs or infrastructure around it are free.
Activity level: Very active · Last code update: 07 Nov 2026 · Created: 24 Nov 2022 · Open issues: 47
Official links: Official Website ↗ · Releases ↗ · Documentation (README) ↗
We never publish installation steps we have not verified — the README above is the source of truth.
⭐ 20957 stars · 🍴 2112 forks · 🐛 47 open issues · 💻 Language: Python · 📄 License: MIT · Default branch: main
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
Port of OpenAI's Whisper model in C/C++
A free, open source, and extensible speech-to-text application that works completely offline.
Privacy-first AI meeting note taker for Windows and macOS. Free, open source (MIT). On-device Whisper/Parakeet transcription and speaker diarization, no meeting bot. Open-source…
Distribute and run LLMs with a single file.