Fish Audio launches S2.1 Pro with support for 83 languages
Fish Audio is pitching S2.1 Pro as its main production model for live voice agents.
The model is available through the Fish Audio API, with a free tier for development and testing under fair-use limits. TestingCatalog says Fish Audio claims about 90 milliseconds to first audio, plus one voice identity that carries across 83 languages. It also supports bracketed delivery cues, multi-speaker dialogue, and short-sample voice cloning. TestingCatalog's note
The model is available through the Fish Audio API, with a free tier for development and testing under fair-use limits. TestingCatalog says Fish Audio claims about 90 milliseconds to first audio, plus one voice identity that carries across 83 languages. It also supports bracketed delivery cues, multi-speaker dialogue, and short-sample voice cloning. TestingCatalog's note
score 6