List your company

Automatic Speech Recognition on Spgispeech (test)

2.17WERSpeech-Hands
2.03042.97273.9154.8573Jan 14, 2026Jan 20, 2026Jan 26, 2026Feb 2, 2026Feb 8, 2026Feb 14, 2026Feb 21, 2026

Evaluation results

MethodLinks
Speech-HandsFramework=Qwen2.5-Omni, External Reference=Canary-1b-v22026.012.17
Speech-HandsFramework=Qwen2.5-Omni, External Reference=Parakeet-tdt-0.6b-v32026.012.26
ReHear2026.022.36
ReHear + Rule2026.022.37
ReHear + Model2026.022.39
ISL2026.022.41
IPL + Rule2026.022.57
IPL2026.022.83
Speech-HandsFramework=Qwen2.5-Omni, External Reference=Whisper-v2-large2026.013.01
Canary-1b-v2Model Type=ASR model2026.013.08
Phi-4-MMModel Type=Omni-LLM2026.013.13
Parakeet-tdt-0.6b-v3Model Type=ASR model2026.013.16
Gemini-2-FlashModel Type=Omni-LLM2026.013.82
Whisper-v2-largeModel Type=ASR model2026.013.94
GER (Cascaded)Framework=Qwen2.5-Omni, External Reference=Whisper-v2-large2026.013.94
GER (Cascaded)Framework=Qwen2.5-Omni, External Reference=Parakeet-tdt-0.6b-v32026.013.98
GER (Cascaded)Framework=Qwen2.5-Omni, External Reference=Canary-1b-v22026.014.02
Qwen2.5_omniModel Type=Omni-LLM2026.014.58
GPT-4o-voiceModel Type=Omni-LLM2026.015.66