Fish-Speech-1.5
About Fish-Speech-1.5
Fish Speech V1.5 is a leading open-source text-to-speech (TTS) model. The model employs an innovative DualAR architecture, featuring a dual autoregressive transformer design. It supports multiple languages, with over 300,000 hours of training data for both English and Chinese, and over 100,000 hours for Japanese. In independent evaluations by TTS Arena, the model performed exceptionally well, with an ELO score of 1339. The model achieved a word error rate (WER) of 3.5% and a character error rate (CER) of 1.2% for English, and a CER of 1.3% for Chinese characters.
Available Serverless
Run queries immediately, pay only for usage
$
15.0
Per 1M UTF-8 bytes
Metadata
Specification
State
Available
Architecture
Calibrated
No
Mixture of Experts
No
Total Parameters
Activated Parameters
Reasoning
No
Precision
FP8
Context length
0K
Max Tokens
Supported Functionality
Serverless
Supported
Serverless LoRA
Not supported
Fine-tuning
Not supported
Embeddings
Not supported
Rerankers
Not supported
Support image input
Not supported
JSON Mode
Not supported
Structured Outputs
Not supported
Tools
Not supported
Fim Completion
Not supported
Chat Prefix Completion
Not supported
Compare with Other Models
See how this model stacks up against others.
IndexTeam
text-to-speech
IndexTTS-2
Release on: Sep 10, 2025
Total Context:
0K
Max output:
Input:
$
/ M UTF-8 bytes
Output:
$
/ M UTF-8 bytes
Fish Audio
text-to-speech
Fish-Speech-1.5
Release on: Nov 29, 2024
Total Context:
0K
Max output:
Input:
$
/ M UTF-8 bytes
Output:
$
/ M UTF-8 bytes
FunAudioLLM
text-to-speech
FunAudioLLM/CosyVoice2-0.5B
Release on: Dec 16, 2024
Total Context:
0K
Max output:
Input:
$
/ M UTF-8 bytes
Output:
$
/ M UTF-8 bytes
Model FAQs: Usage, Deployment
Learn how to use, fine-tune, and deploy this model with ease.
