Stories by xenova

Kokoro WebGPU: Real-time text-to-speech 100% locally in the browser

Transformers.js v3: WebGPU Support, New Models and Tasks, and More

SAM 2: Segment Anything in Images and Videos

PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding

Mamba: New SSM arch with linear-time scaling that outperforms Transformers

Count tokens used by GPT-4 and Llama for large texts (> 50k characters)

Making real-time ML-powered web games with Transformers.js

MMS: Scaling Speech Technology to 1000 languages demo

Whisper Web: ML-powered speech recognition in the browser