Speech & Audio

Apriel Voice 27B Preview is an open speech-language model built on Gemma 4 26B-A4B that uses a two-stage speech-adaptation recipe to unify speech understanding, reasoning, conversation, and tool calls while retaining text and image inputs.

CoSTA decouples an SDM's generation and synthesis token spaces by extending the generation vocabulary with BPE-compressed speech units, adapting the model, and deterministically expanding generated units back before the unchanged decoder, yielding about 1.7× speech-token generation speed.

Falcon is a single-pass causal audio-visual target speaker extractor paired with the RealSSA benchmark, achieving strong extraction quality with much lower parameter, computation, and latency costs than causal AV-TFGridNet.