3 ms·
Flashlabs releases the world’s first open-source voice cloning model
- phs318u 9mo agoThe GitHub repo URL returns a 404. https://github.com/FlashLabs-AI-Corp/FlashLabs-Chroma https://github.com/FlashLabs-AI-Corp/FlashLabs-Chroma
- kuandriy 9mo agoThe end-to-end speech-to-speech claim is interesting, especially avoiding the ASR→LLM→TTS pipeline, which is where most latency and error compounding happens.