Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-form Audio Quality Description on QualiSpeech
Loading...
51
ROUGE-L
LLM-judge dim.-wise
10.44
20.97
31.5
42.03
Mar 10, 2026
ROUGE-L
Correlation Coefficient
Updated 1mo ago
Evaluation Results
Method
Method
Links
ROUGE-L
Correlation Coefficient
LLM-judge dim.-wise
Reasoning strategy=LLM...
2026.03
51
0.83
Acc.+Sem. dim.-wise
Reasoning strategy=Acc...
2026.03
47
0.8
SQ-LLM
Type=Baseline
2026.03
45
0.82
QualiSpeech-FT
Type=Baseline
2026.03
40
0.78
QualiSpeech-FT-SALMONN
Type=Baseline
2026.03
38
0.75
x (Calibration-only)
Strategy=Calibration-o...
2026.03
12
0.19
Feedback
Search any
task
Search any
task