od Fish Audio
Fish Audio Drama 3
Directed performances — plain-language [directions], paired word tags, multi-speaker scenes
fish-audio-drama-3Convert text to speech with this model.
/api/v1/audio/speech{
"model": "fish-audio-drama-3",
"input": "Welcome to Kunya. This is a text-to-speech demonstration.",
"voice": "alloy"
}curl -X POST https://kunya.ai/api/v1/audio/speech \
-H "Authorization: Bearer kunya_YOUR_API_KEY" \
-H "Content-Type: application/json"
-d '{"model":"fish-audio-drama-3","input":"Welcome to Kunya. This is a text-to-speech demonstration.","voice":"alloy"}'Fish Audio
Production TTS — 83 languages, [bracket] emotions, multi-speaker dialogue
Dokumentacja APIFish Audio
Same model as S2.1 Pro for dev/test — no TTFA/DPA guarantees; fair-use limits
Dokumentacja APIOpenAI
Expressive text-to-speech — steerable tone, emotion, and accent
Dokumentacja API