Vosk

Health v2 · maintenance only

Adoption not part of the score

15077 stars · 1749 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Vosk is an offline open source speech recognition toolkit supporting 20+ languages with small models and a streaming API. It provides bindings for Python, Java, Node.js, C#, and more, scaling from Raspberry Pi and mobile devices to server clusters.

Use cases

• transcribe audio to text offline
• add speech recognition to an android app
• create subtitles for movies
• transcribe lectures and interviews
• build a voice-controlled assistant or chatbot
• run speech-to-text on a raspberry pi
• identify speakers in audio recordings

When to choose

• you need fully offline, privacy-preserving speech recognition
• you need small models that run on mobile or embedded devices
• you need streaming, low-latency transcription
• you need bindings across many programming languages

When to avoid

• you need the highest accuracy cloud-grade ASR and have internet access
• you need languages not covered by available Vosk models
• you need built-in text-to-speech rather than recognition

Facets

library · maturity stable

speech-recognition machine-learning audio-processing sdk speech-processing machine-learning privacy python cross-platform cli embedded asr speech-to-text offline kaldi streaming-transcription speaker-identification raspberry-pi self-hosted natural-language-processing android ios nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="alphacep/vosk-api")