Fish Audio launches Drama 3: direct AI voice lines with natural-language emotion and sound cues
rohanpaul_ai · x · 2026-10-14
- Fish Audio released a preview of Drama 3, a voice model built around "directing the performance, not just the words": users describe how a line should feel in natural language, add sounds, emotions, or pauses, and control exactly which words change.
- The demo page shows five ways to direct the same line: no direction, natural-language description (e.g. "exasperated, like a very disappointed uncle"), inserting sound effects, picking specific words, and any-language switching. Directions mix freely in one line as plain, editable, reusable text.
- All samples are generated by Drama 3, including a licensed Uncle Roger voice demo roasting kitchen crimes like ketchup fried rice.
- Free trial available; link shared by rohanpaulai.
Related event: Fish Audio Open-Sources Drama 3 TTS Model, Discloses $52M Seed Round(4 posts)→
More from Multimodal
- Gemini 4 Argon recreates Manhattan photo as oil painting via multi-pass canvas renderer — _philschmid · 2026-10-14
- Hollywood embraces AI tools despite doomsday fears, Variety reports from studio stages — DavidmComfort · 2026-10-14
- Early user teases Gemini 4 Argon: impressive 3D modeling and spatial reasoning — JMateosGarcia · 2026-10-14
- Netflix's Eyeline Studios details its AI 'hybrid production' pipeline: relighting, location swaps and camera repositioning after the shoot — DavidmComfort · 2026-10-14
- fal launches Reframe: change any video's aspect ratio without cropping or reshooting — jfischoff · 2026-10-14
- CineConIA builds free ComfyUI nodes for AI filmmaking camera control — CineConIA · 2026-10-14