2026-11-02 · America/Los_Angeles · 论文 · #11

Apriel Voice 27B Preview: An Open Model for Speech, Reasoning, and Tool Use

A capable voice assistant must do more than recognize words; it must understand requests, reason over context, sustain conversations, and act through tools. We present Apriel Voice 27B Preview, an open speech-language model that unifies these capabilities within a shared reasoning backbone while retaining text and image input modalities. Built on the post-trained Gemma 4 26B-A4B checkpoint, Apriel Voice directly processes speech and generates both responses and tool calls without requiring a separate dialogue model. We introduce a two-stage training procedure to adapt the model to speech.…

热度 39.4 / 100；排名与评分保留该期记录。