Drama 3
#6 in Spraaksynthese (TTS)Fish Audio · v3 · siet 13. Oktober 2026 · 4× · tolest 17. Okt. 2026
56
Momentum
Drama 3 is a text-to-speech model from Fish Audio built for voice direction. It was released on October 13, 2026 and is available via fish.audio, the Fish Audio app (iOS/Android) and the API. The weights are published as open weights on Hugging Face under the Fish Audio Community License v1.0.
Momentum-Verloop
20.07.18.10.
Features
| Latency | 88.9 ms median time to first audio (1 stream, NVIDIA H200, self-reported) |
| License | Fish Audio Community License v1.0 (open weights, not open source): free for research/non-commercial and commercial use under $500,000 annual revenue; hosting as a service needs a separate license |
| Platform | fish.audio (web), Fish Audio app (iOS, Android), API (POST /v1/tts, model: drama-3), open weights on Hugging Face |
| Price | API: $10 per 1M characters (launch until Nov 12, 2026), list price $50 per 1M characters |
| Release Date | October 13, 2026 |
| Languages | 83 languages |
| Voice Cloning | Yes (voice cloning; license bans cloning a real person's voice without consent) |