Interfaze Ships diffusion-gemma-asr-small, an Open-Source Diffusion ASR Model Transcribing Six Languages via DiffusionGemma’s Parallel Denoising Decoder

Interfaze, a young YCs startup, has open-sourced a new speech recognition model. It is called . The model transcribes audio through a diffusion decoder, not an autoregressive one. It is described as the first multilingual audio diffusion ASR model. One adapter handles six languages. The research team trained only about 42M parameters on top of a frozen 26B backbone. That is roughly 0.16% of the models weights. Here two terms matter up front. Autoregressive models generate text one token at a time. Diffusion models refine all tokens in parallel. This model uses the diffusion approach for speech-to-text. TL;DR Claimed by the Interfaze team, to be the first open-source multilingual diffusion ASR: six languages from a single ~42M-parameter adapter. Transcribes via DiffusionGemmas diffusion decoder using uniform, random-token diffusion, not the absorbing scheme. Transcription cost scales with denoising steps, not transcript length. Leads diffusion peers on LibriSpeech (6.6% WER vs Whisfusions 8.3%) but trails autoregressive Whisper. The adapter ships under Apache-2.0; DiffusionGemma (Gemma terms) and whisper-small (MIT) load separately. What is diffusion-gemma-asr-small? is an audio-native ASR model. It converts speech to text using a discrete diffusion decoder. That decoder belongs to DiffusionGemma, Googles 26B mixture-of-experts model. DiffusionGemma activates 4B parameters, using 128 experts with top-8 routing. It generates text by discrete diffusion instead of autoregression. The diffusion detail is specific. Most diffusion LLMs use an absorbing scheme. DiffusionGemma uses uniform, random-token diffusion instead. It fills a fixed-length canvas with random vocabulary tokens. Each step keeps confident predictions and re-randomizes the rest. After a few steps the noise anneals into text. Interfaze added audio to this text-only model. Out of the box, DiffusionGemma takes text, images, and video. It does not take audio. The repo ships only the trained adapter, about 42M