News7 min
A Diffusion Model for Speech-to-Text Just Went Open Source. Why Ad Captions Care
Interfaze open-sourced a speech-to-text model that transcribes by diffusion instead of word-by-word decoding: six languages from a tiny adapter on frozen models. Here is what the launch actually claims, why transcription quality sets the ceiling on auto-captions, and what captioning a video costs in Novoads today.
