Qwen3 TTS lands on Together AI: 9 voices, 10 languages, first audio in 97ms
togethercompute · x · 2026-11-21
Qwen3 TTS is now live on Together AI's Serverless Inference and dedicated endpoints at $20 per 1M characters. The open model separates who speaks from how they speak: 9 preset voices whose tone and emotion can be changed mid-stream via plain-language instructions like "excited and upbeat," with each voice staying consistent across 10 languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, Italian). First audio arrives in as little as 97ms, with HTTP streaming and a WebSocket API for incremental text. All 9 voices are in VoiceFinder alongside 850+ others for side-by-side comparison on your own text.
Related event: Qwen3 TTS Lands on Together AI with 9 Voices and 10 Languages(2 posts)→
More from Multimodal
- Ex-TensorFlow early member founds TensorScale, betting world models will beat LLMs — TerryTangYuan · 2026-11-21
- Cinematographer Tests Minimax's Cinestill 800T Emulation for a Blade Runner Look in Neon Rain — Purple-Let4907 · 2026-11-21
- 15-year electronic producer built an agentic DAW with Claude and made a full album — yawnxyz · 2026-11-21
- Self-hosted cloud GPUs beat Higgsfield and Runway APIs on cost and flexibility — Al_Grigor · 2026-11-21
- Gaussian splat top-down scan shows how Fable plans object placement in pixel apartment — XRarchitect · 2026-11-21
- Pixel replica of a real apartment built from Atlas gaussian splats with Fable 5.2 — XRarchitect · 2026-11-21