Fish Audio launches S2.1 Pro with support for 83 languages

Fish Audio has launched S2.1 Pro as its recommended production voice model.

It is built for real-time conversational speech with roughly 90 ms time to first audio.

The model supports 83 languages, multi-speaker dialogue, emotional control, and short-sample voice cloning.

It is available via the Fish Audio API and a free testing tier, with MCP and agent-skill integration for workflows.