← All models

Fish Audio S1

Audio Available Full comparison ↗

fish-audio/s1 · by Fish Audio · neural-tts

Pricing — 1 offering(s)

Fish Audio

Text characters

  • $15.00 / 1M characters Current 2026-07-30 → present
    Fish Audio's native billing unit is UTF-8 BYTES, not characters — the pricing page states "$15.00 / M UTF-8…

    Fish Audio's native billing unit is UTF-8 BYTES, not characters — the pricing page states "$15.00 / M UTF-8 bytes". For English and other Latin-script text 1 character = 1 byte, so the rate is equivalent; for CJK and other multi-byte scripts, actual cost per character can run up to 3-4x this rate. Recorded here as per_1m_characters (closest schema unit) since the registry has no byte-based unit; effective_from is the date this entry was sourced, not a confirmed price-change date.

Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.

See how Fish Audio S1 fits into a cost-aware routing setup

See how →

Capability profile

How Fish Audio S1 rates across core capability dimensions, with the task-level evidence behind each rating.

voice naturalness Strong

0.8% WER / 0.4% CER per Fish Audio's own figures; #1 ranking on the independent TTS-Arena2 leaderboard at time of writing — one of the stronger independently-corroborated naturalness signals in this registry. Cross-checked 2026-08-19 (SCO-462 sweep) against a second, separate independent arena: the Artificial Analysis Text to Speech Arena (artificialanalysis.ai/text-to-speech/leaderboard/provider-voice) lists this model as "OpenAudio S1" — confirmed to be the same model (OpenAudio is Fish Audio's research-lab brand for this exact release, per openaudio.com/blogs/s1) — at #42 of 98 voices tracked, Elo 1079. This is a more modest, mid-table result than the TTS-Arena2 "#1" claim — the two are different arenas with different voter pools and snapshot dates, so this doesn't contradict the TTS-Arena2 figure, but it's a meaningfully less flattering independent data point worth recording alongside it rather than only keeping the stronger claim.

language support Moderate

13 languages — narrower than S2 Pro/S2.1 Pro (80+/83) and below ElevenLabs Multilingual v2 (29) or Eleven v3 (70+).

voice variety Strong

Voice cloning from as little as 10-15 seconds of reference audio, plus access to Fish Audio's community voice library (2M+ voices claimed platform-wide, not S1-specific).

streaming latency Unknown

No S1-specific latency figures published. S2 Pro/S2.1 Pro carry an explicit 100ms time-to-first-audio guarantee that S1 does not — Fish Audio's docs don't state whether S1 shares comparable latency.

cloning support Strong

Zero-shot voice cloning from ~10-15 seconds of reference audio, per Fish Audio's platform-wide cloning claim.

Benchmarks

Benchmark Score Config Source
Word error rate (vendor-reported) 0.8 % Fish Audio's own reported WER. Not independently re-measured by Modelglass. source ↗
TTS-Arena2 ranking (independent, community leaderboard) — Ranked #1 at time of writing per Fish Audio's own docs — an independent/community leaderboard, not a vendor-run benchmark, though the ranking snapshot itself wasn't independently re-verified by Modelglass. —

Operator guidance

Fish Audio's own docs label S1 "Previous Model" and recommend s2.1-pro for all new projects — S1 is kept available only "for existing integrations," priced identically to S2 Pro/S2.1 Pro. Use S2.1 Pro instead for anything new; S1 is documented here for operators already integrated against it.

Use cases

Limitations

Citations