​
Available Models

Fish Audio offers state-of-the-art text-to-speech models optimized for different use cases and performance requirements.

​
Current Model

s1

OpenAudio S1 - Our flagship model with industry-leading quality
  • 4 billion parameters
  • 0.008 WER (0.8% word error rate)
  • Best performance and naturalness
  • Full emotional control capabilities

​
Legacy Models

speech-1.6

Fish Speech 1.6 - Previous generation model
  • Stable and production-tested
  • Good performance for standard use cases
  • Basic emotional control

speech-1.5

Fish Speech 1.5 - Earlier generation model
  • Reliable performance
  • Limited to basic emotions
  • Lower resource requirements
We recommend using s1 for all new projects to access the latest capabilities and performance improvements. Legacy models remain available for existing integrations.

​
Model Specifications

​
OpenAudio S1

SpecificationValue
Parameters4B
Context Length8,192 tokens
Languages30+ languages
Emotions48+ expressions
Latency~200ms
Max Audio Length30 minutes
Streaming✅ Yes

​
Performance Metrics

  • Word Error Rate (WER): 0.008 (0.8%)
  • Character Error Rate (CER): 0.004 (0.4%)
  • Real-time Factor: ~1:7 on standard hardware
  • TTS-Arena2 Ranking: #1 worldwide

​
Supported Languages

Fish Audio models support text-to-speech generation in 30+ languages with full emotional expression capabilities.

​
Primary Languages

English (en), Chinese (zh), Japanese (ja), Korean (ko),
Spanish (es), French (fr), German (de), Portuguese (pt),
Italian (it), Russian (ru), Arabic (ar), Hindi (hi)

​
Additional Languages

Dutch (nl), Polish (pl), Turkish (tr), Swedish (sv),
Norwegian (no), Danish (da), Finnish (fi), Czech (cs),
Hungarian (hu), Romanian (ro), Bulgarian (bg), Greek (el),
Hebrew (he), Thai (th), Vietnamese (vi), Indonesian (id),
Malay (ms), Filipino (tl), Ukrainian (uk)
Language detection is automatic - simply provide text in your target language. You can also specify the language explicitly using the language parameter in your API request.

​
Voice Styles and Emotions

OpenAudio models support 64+ emotional expressions and voice styles that can be controlled through text markers in your input.

​
Basic Emotions (24 expressions)

(angry) (sad) (excited) (surprised) (satisfied) (delighted)
(scared) (worried) (upset) (nervous) (frustrated) (depressed)
(empathetic) (embarrassed) (disgusted) (moved) (proud) (relaxed)
(grateful) (confident) (interested) (curious) (confused) (joyful)

​
Advanced Emotions (25 expressions)

(disdainful) (unhappy) (anxious) (hysterical) (indifferent)
(impatient) (guilty) (scornful) (panicked) (furious) (reluctant)
(keen) (disapproving) (negative) (denying) (astonished) (serious)
(sarcastic) (conciliative) (comforting) (sincere) (sneering)
(hesitating) (yielding) (painful) (awkward) (amused)

​
Tone Markers (5 expressions)

(in a hurry tone) (shouting) (screaming) (whispering) (soft tone)

​
Audio Effects (10 expressions)

(laughing) (chuckling) (sobbing) (crying loudly) (sighing)
(panting) (groaning) (crowd laughing) (background laughter) (audience laughing)
You can also use natural expressions like “Ha,ha,ha” for laughter. Experiment with combinations to achieve the perfect emotional tone for your application.