Flow - 4x faster than typing
Mini toolsS2.1 Pro
By Fish Audio
Model family: Fish
S2.1 Pro is a production text to speech model and an improved version of S2 Pro, with higher quality, lower latency, and greater throughput. It synthesizes speech in 83 languages using automatic language detection and treats bracket tags such as [whispers sweetly] or [laughing nervously] as natural language cues, so expression is not limited to a fixed set. It supports multi-speaker dialogue and instant voice cloning from a short reference audio clip plus its transcript. Output is available as mp3, wav, pcm, or opus, with adjustable speed, prosody, bitrate, and sampling parameters. Real-time streaming over WebSocket delivers roughly 70 to 90 ms time to first audio, suiting voice agents, audiobooks and narration, IVR, game dialogue, and multilingual applications. It reports a 61 percent win rate against S2 Pro in listening evaluations.
New
Audio
Overview
S2.1 Pro is a neural text to speech model that converts text into speech across 83 languages with automatic language detection. It supports inline bracket tags for emotion and paralinguistic control, multi-speaker dialogue, and instant voice cloning from a reference audio sample. It streams audio in real time with about 70 to 90 ms time to first audio and outputs mp3, wav, pcm, or opus.
Last updated: July 13, 2026
