Tencent Hunyuan open-sources unified speech model AuK

Tencent Hunyuan released AuK, a fully open-sourced 1.5B speech foundation model that unifies generation and editing via natural language instructions, covering TTS, content/acoustic/paralinguistic editing, enhancement and separation. A distilled Flash variant is 4.5x faster.

2026-09-09 ~ 2026-09-10 · 3 related posts

• Tencent Hunyuan open-sources AuK, a foundational model unifying speech generation and editing — Tencent-Hunyuan · 2026-09-09
• Tencent releases AuK, a unified speech generation and editing model with a 4.5x faster Flash version — _akhaliq · 2026-09-10

1 near-duplicate retellings: realmrfakename