Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Utterance-level Pairwise Preference Judgement on UniSRM-BENCH T1
Loading...
65.06
Accuracy
UniSRM
47.3072
51.9161
56.525
61.1339
May 22, 2026
Accuracy
Updated 9d ago
Evaluation Results
Method
Method
Links
Accuracy
UniSRM
Model Category=Propose...
2026.05
65.06
GPT-4o-Audio
Model Category=Proprie...
2026.05
61.04
Gemini-2.5-Pro
Model Category=Proprie...
2026.05
60.67
Gemini-2.5-Flash
Model Category=Proprie...
2026.05
60.44
WER
Type=Objective Metric
2026.05
59.24
SpeechJudge
Model Category=Open-So...
2026.05
57.2
Kimi-Audio-7B
Model Category=Open-So...
2026.05
52.81
Qwen2.5-Omni-7B
Model Category=Open-So...
2026.05
51.2
MiMo-Audio-7B
Model Category=Open-So...
2026.05
50.4
UTMOS
Type=Objective Metric
2026.05
50.2
DNSMOS
Type=Objective Metric
2026.05
49.8
SIM
Type=Objective Metric
2026.05
47.99
Feedback
Search any
task
Search any
task