vosk-browser and voskJs
These are complementary tools serving different runtime environments: the browser-based WebAssembly implementation (A) enables client-side speech recognition in web applications, while the Node.js wrapper (B) provides server-side ASR capabilities, allowing developers to choose between in-browser or backend processing depending on their architecture needs.
About vosk-browser
ccoreilly/vosk-browser
A speech recognition library running in the browser thanks to a WebAssembly build of Vosk
Explicitly targets WebWorker execution to avoid blocking the main thread, with support for real-time microphone input or pre-recorded audio files across 13 languages. Integrates with the Web Audio API for audio processing and provides both npm and CDN distribution, offering event-based result callbacks for partial and final recognition outputs. The library wraps an updated Vosk WebAssembly build with a JavaScript API, handling model loading and audio pipeline management.
About voskJs
solyarisoftware/voskJs
Vosk ASR offline engine API for NodeJs developers. With a simple HTTP ASR server.
Provides both sentence-based and event-driven streaming APIs for speech recognition, emitting `partial`, `endOfSpeech`, and `final` events as audio is processed. Built as a functional wrapper around Vosk's DNN-HMM architecture (based on Kaldi), supporting low-latency offline transcription (<500ms) with multi-threaded server deployments and optional grammar constraints. Includes CLI tools (`voskjs`, `voskjshttp`) and integrates with frameworks like Rhasspy for voice assistant applications.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work