Client-side ASR uses compact models — faster-whisper compiled to WebAssembly, for example — to transcribe on the device. It protects privacy and cuts cost, at the price of fitting within limited local compute, often aided by WebGPU.
Definition
Running speech recognition in the user's browser or device instead of a server, keeping audio local and removing per-minute cloud costs.
Client-side ASR uses compact models — faster-whisper compiled to WebAssembly, for example — to transcribe on the device. It protects privacy and cuts cost, at the price of fitting within limited local compute, often aided by WebGPU.
Also known as
in-browser ASR, on-device transcription
Specialist software house for video, real-time and AI products. Founded 2005. 50 in-house engineers.
