alexkroman/tiny-audio

A speech-to-text system you can train for $25. A frozen speech encoder and a Qwen LLM joined by a small trained projector, with word timestamps, speaker diarization, a batched HTTP server, a live demo, and a free course.

10 best points · 0 comments · 1 Hacker News post · 179 GitHub stars · View on GitHub

Hacker News discussions

• tiny-audio, nanoGPT for speech-to-text (10 points, 0 comments, by mazesmazes, 2026-10-09)