▸_Vipul Srivastava

Three modalities, still no server

The other playgrounds need WebGPU. These don't. transformers.js (ONNX Runtime Web) runs real speech, vision, and embedding models on your CPU via WebAssembly — so speech-to-text, zero-shot image understanding, and semantic search all work in any modern browser, with nothing leaving the tab.

Runs on your CPU — no GPU required
Models run in WebAssembly (optionally WebGPU-accelerated). First use downloads the weights once into your browser cache — Whisper-tiny ~40 MB, CLIP ~90 MB, MiniLM ~30 MB — then everything runs offline.

Whisper transcribes audio, locally.

Play the built-in sample, upload an audio file, or record from your mic. It's decoded to 16 kHz mono in the browser and transcribed by Xenova/whisper-tiny.en. The audio never leaves the tab.

automatic-speech-recognition · Xenova/whisper-tiny.en · ~40 MB

Ready.
The transcript will appear here.

CLIP scores how well an image matches each label.

Zero-shot image classification: no training for these labels — Xenova/clip-vit-base-patch32 embeds the image and each candidate phrase into the same space and ranks them by similarity. Pick a sample or upload your own, edit the labels, and run.

zero-shot-image-classification · Xenova/clip-vit-base-patch32 · ~90 MB

sample: person sample: scene
selected image

Candidate labels

One phrase per line. These are the classes CLIP will rank the image against.

Ready.