Our new X account is live! Follow @wizwand_team for updates

Speech Recognition on Common Voice EN

6.76WER

Step-Audio2-Mini

Updated 4d ago

Evaluation Results

Method	Links
Step-Audio2-Mini 2025.12		6.76
Audio-Flamingo 2025.12		7.4
Qwen-TTS-Tokenizer-25Hz (Stage 1) 2026.01		7.51
Fun-Audio-Chat 2025.12		7.79
Fun-Audio-Chat 2025.12		8.88
Kimi-Audio 2025.12		10.31
Qwen-TTS-Tokenizer-25Hz (Stage 2) 2026.01		10.4
S3 Tokenizer(FSQ) 2026.01		10.67
S3 Tokenizer(VQ) 2026.01		11.56
S3 Tokenizer(VQ) 2026.01		12.06
MiMo-Audio 2025.12		62.05