UniSRM-BENCH
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| utterance-level pairwise preference judgement | UniSRM-BENCH T1 | Accuracy65.06 | 12 | |
| multi-turn dialogue speech evaluation | UniSRM-BENCH T4 | Accuracy88.89 | 10 | |
| scenario-aware style consistency preference (Chinese) | UniSRM-BENCH T3-Zh | Accuracy91.3 | 10 | |
| scenario-aware style consistency preference (English) | UniSRM-BENCH T3-En | Accuracy85.61 | 10 | |
| fine-grained speech quality scoring | UniSRM-BENCH T2 | PCC0.551 | 9 |
Showing 5 of 5 rows