Vosk is an offline open source speech recognition toolkit. It enables speech recognition models for 12 languages and dialects - English, Indian English, German, French, Spanish, Portuguese, Chinese, Russian, Turkish, Vietnamese, Italian, Dutch.
Vosk models are small (50 Mb) but provide continuous large vocabulary transcription, zero-latency response with streaming API, reconfigurable vocabulary and speaker identification.
Speech recognition bindings implemented for various programming languages like Python, Java, Node.JS, C#, C++ and others.
Vosk supplies speech recognition for chatbots, smart home appliances, virtual assistants. It can also create subtitles for movies, transcription for lectures and interviews.
Vosk scales from small devices like Raspberry Pi or Android smartphone to big clusters.
For installation instructions, examples and documentation visit Vosk Website. See also our project on Github.