Interfaze Ships diffusion-gemma-asr-small, an Open-Source Diffusion ASR Model Writes Six Languages with DiffusionGemma’s Parallel Denoising Decoder
Interfaze, a young YC startup, has opened the source for a new speech recognition model. It is called diffusion-gemma-asr-small. The model encodes noise with a diffusion decoder, not an autoregressive one. It is described as the first ASR model for multilingual audio streaming. One adapter handles six languages. The research team trained only 42M parameters … Read more