hjLabs AI Playground
Starting the on-device runtime…
Whisper Large v3 Turbo is an 809M-parameter ONNX model that transcribes speech from an audio file or a live microphone into text with timestamps. On this page it runs entirely inside your browser: the 983 MB download comes straight from the Hugging Face repository “onnx-community/whisper-large-v3-turbo” into your browser cache, and every inference after that executes on your own GPU through WebGPU. This is a serious download — do it once on a connection you are not paying by the megabyte for. Once it is in the Cache Storage API it stays there, and subsequent visits load it from disk in seconds. The model is multilingual, so it is not restricted to English input.
Near state-of-the-art transcription in the browser, if you can spare ~1 GB of cache and a slow first run. In practice people reach for a model like this for transcribing interviews, meetings, voice notes and video soundtracks without uploading the audio anywhere.
Available precisions: q8 (8-bit) and q4 (4-bit). q8 comes first because it roughly halves the download at a quality cost you will struggle to measure on this class of model; fp32 is there when you want the reference numbers. Hugging Face records about 16k downloads a month. It is a specialist choice rather than a default, which is exactly why it is worth trying when the popular model gets your particular input wrong.
Because there is no inference server behind this page, there is nothing to rate-limit and nothing to bill: the only cost of running Whisper Large v3 Turbo here is the one-off 983 MB download, and once that is cached the page keeps working with the network switched off.
whisper large v3 turbowhisper large v3 turbo onnxwhisper large v3 turbo browserrun whisper large v3 turbo locallyonnx-community/whisper-large-v3-turbospeech to text modelwhisper large v3 turbo webgpuwhisper large v3 turbo download sizefree speech to text modeltransformers.js model
About 983 MB for the q8 build, including the tokenizer and config files the runtime also has to fetch — this is the whole payload, not just the weights file. It is downloaded once and then served from the browser Cache Storage API.
Yes, through the WebAssembly fallback, which is slower but works anywhere. WebGPU (Chrome or Edge 113+, Safari 26+) is what makes it feel interactive. On WASM, prefer the smaller quantized build.
No. onnx-community/whisper-large-v3-turbo is a public repository, so your browser fetches the ONNX files directly over HTTPS. There is no account, no key and no server in between.
No. The model weights are downloaded from Hugging Face to your browser once, and every inference after that runs on your own GPU through WebGPU. Your text, images and audio are never sent anywhere — hjLabs.in has no inference server and no way to see your input.