腾讯混元开源语音生成与编辑基础模型 AuK，同步发布 4 步推理的 AuK-Flash
🚀 AuK is officially here. Nano banana🍌 for audio
腾讯混元正式发布开源基础模型 AuK，面向统一的语音生成与编辑，采用自然语言指令加参考音频的单一接口，支持零样本 TTS、指令控制生成、内容编辑、耳语转换、去口音、音色风格情感编辑、语速音高控制、增强降噪、多说话人分离与音乐分离。同时发布 AuK-Flash，采用 4 步推理，在匹配条件下约快 4.5 倍。代码、权重与 demo 均已上线。
正文
当前显示原文，中文全文译文尚未就绪。
🚀 AuK is officially here. Nano banana🍌 for audio
An open-source foundation model for unified speech generation and editing. Natural-language instructions + reference audio. One interface. Zero-shot TTS. Instruction-controlled generation. Content editing. Whisper-conversion. De-accent. Timbre/style/emotion edit. Speed/Pitch control. Enhancement, denoising, multi-speaker and music separation. Also releasing AuK-Flash: 4-step inference. ~4.5× faster under matched conditions. Code, weights, and demo are live. Try it and share your feedback.