SOTA Automatic Speech Recognition on GigaSpeech (test) and PapersWithCode

9.78WER

Phi-4-MM

Updated 3d ago

Evaluation Results

Method	Links
Phi-4-MM 2026.01		9.78
Pruned transducer w/ CR-CTC, Zipformer-XL 2024.10		10.03
CR-CTC/AED, Zipformer-XL 2024.10		10.07
Pruned transducer, Zipformer-XL 2024.10		10.2
Pruned transducer, Zipformer-L 2024.10		10.28
CR-CTC, Zipformer-XL 2024.10		10.28
Index-ASR 2025.12		10.29
CTC/AED, Zipformer-XL 2024.10		10.33
CTC/AED, Zipformer-L 2024.10		10.38
CR-CTC, Zipformer-L 2024.10		10.41
Pruned transducer, Zipformer-M 2024.10		10.42
CTC/AED, E-Branchformer in ESPnet 2024.10		10.5
CTC/AED, Conformer in Wenet 2024.10		10.6
CTC/AED, Zipformer-M 2024.10		10.61
Gemini-2-Flash 2026.01		10.71
CR-CTC, Zipformer-M 2024.10		10.72
CTC/AED, Conformer in ESPnet 2024.10		10.8
CTC, Zipformer-XL 2024.10		10.87
Speech-Hands 2026.01		10.87
Pruned transducer, Zipformer-S 2024.10		10.94
Speech-Hands 2026.01		11.1
CTC, Zipformer-L 2024.10		11.16
Qwen2.5_omni 2026.01		11.26
CTC, Zipformer-M 2024.10		11.27
FireRedASR 2025.12		11.33
CTC/AED, Zipformer-S 2024.10		11.39
Whisper-v2-large 2026.01		11.45
CR-CTC, Zipformer-S 2024.10		11.58
Canary-1b-v2 2026.01		11.66
CTC, Zipformer-S 2024.10		11.95
GER (Cascaded) 2026.01		12.1
GER (Cascaded) 2026.01		12.15
Parakeet-tdt-0.6b-v3 2026.01		12.24
CTC/AED, Transformer 2024.10		12.3
Speech-Hands 2026.01		12.37
GER (Cascaded) 2026.01		12.43
Step-Audio 2 mini 2025.12		13.19
GPT-4o-voice 2026.01		13.64
Whisper 2025.12		14.45
Kimi-Audio 2025.12		16.06