Suno launches Speech beta, the first audio model generating voice with matching music
suno · x · 2026-10-02
Suno has opened Speech (beta) to all users, calling it the first audio model that generates spoken audio and matching background music as one cohesive track. Users type a text and describe the desired voice and musical style; the team demos use cases like dramatic readings of friends' messages, epic-scored voice notes, meditations and bedtime stories. The company admits it's rough: British accents sometimes drift Australian and dramatic pauses get very dramatic. Available now via app update.
More from Multimodal
- AI-generated 'SCP: UNIVERSE' video showcase — imjustnewatai · 2026-10-02
- SageAttention up to 1.96x faster causal attention on RX 9070 XT via hand-written HIP fp8 kernel — Familiar_Worry332 · 2026-10-02
- Dev Redoes All SNES Game Sprites with Scenario MCP, Ships Updated Cartridge — AIandDesign · 2026-10-02
- AI Short 'Beauty and the Beach': Full Pipeline from Midjourney V8.2 to Seedance 2.5 — Kyrannio · 2026-10-02
- Tutorial: integrated dataset prep and LoRA training inside Comfy — no3us · 2026-10-02
- Suno confirms one-pass speech + background music generation that ducks and swells around the voice — suno · 2026-10-02