fishaudio / s1-mini

FishAudio S1

FishAudio S1 is a leading text-to-speech (TTS) model trained on more than 2 million hours of audio data in multiple languages.

Supported languages:

• English (en)
• Chinese (zh)
• Japanese (ja)
• German (de)
• French (fr)
• Spanish (es)
• Korean (ko)
• Arabic (ar)
• Russian (ru)
• Dutch (nl)
• Italian (it)
• Polish (pl)
• Portuguese (pt)

Please refer to Fish Speech Github for more info. Demo available at Fish Audio Playground . Visit the Fish Audio website for blog & tech report.

Emotion and Tone Support

FishAudio S1 supports a variety of emotional, tone, and special markers to enhance speech synthesis:

1. Emotional markers: (angry) (sad) (disdainful) (excited) (surprised) (satisfied) (unhappy) (anxious) (hysterical) (delighted) (scared) (worried) (indifferent) (upset) (impatient) (nervous) (guilty) (scornful) (frustrated) (depressed) (panicked) (furious) (empathetic) (embarrassed) (reluctant) (disgusted) (keen) (moved) (proud) (relaxed) (grateful) (confident) (interested) (curious) (confused) (joyful) (disapproving) (negative) (denying) (astonished) (serious) (sarcastic) (conciliative) (comforting) (sincere) (sneering) (hesitating) (yielding) (painful) (awkward) (amused)

2. Tone markers: (in a hurry tone) (shouting) (screaming) (whispering) (soft tone)

3. Special markers: (laughing) (chuckling) (sobbing) (crying loudly) (sighing) (panting) (groaning) (crowd laughing) (background laughter) (audience laughing)

Special markers with corresponding onomatopoeia:

• Laughing: Ha,ha,ha
• Chuckling: Hmm,hmm

Model Variants and Performance

FishAudio S1 includes the following models:

• S1 (4B, proprietary): The full-sized model.
• S1-mini (0.5B): A distilled version of S1.

Both S1 and S1-mini incorporate online Reinforcement Learning from Human Feedback (RLHF).

Seed TTS Eval Metrics (English, auto eval, based on OpenAI gpt-4o-transcribe, speaker distance using Revai/pyannote-wespeaker-voxceleb-resnet34-LM):

• S1:
  • WER (Word Error Rate): 0.008
  • CER (Character Error Rate): 0.004
  • Distance: 0.332
• S1-mini:
  • WER (Word Error Rate): 0.011
  • CER (Character Error Rate): 0.005
  • Distance: 0.380

License

This model is permissively licensed under the CC-BY-NC-SA-4.0 license.