URGENT-PK: Perceptually-Aligned Ranking Model Designed for Speech Enhancement Competition
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912457949184000 |
|---|---|
| author | Wang, Jiahe Li, Chenda Wang, Wei Zhang, Wangyou Cornell, Samuele Sach, Marvin Scheibler, Robin Saijo, Kohei Fu, Yihui Ni, Zhaoheng Kumar, Anurag Fingscheidt, Tim Watanabe, Shinji Qian, Yanmin |
| author_facet | Wang, Jiahe Li, Chenda Wang, Wei Zhang, Wangyou Cornell, Samuele Sach, Marvin Scheibler, Robin Saijo, Kohei Fu, Yihui Ni, Zhaoheng Kumar, Anurag Fingscheidt, Tim Watanabe, Shinji Qian, Yanmin |
| contents | The Mean Opinion Score (MOS) is fundamental to speech quality assessment. However, its acquisition requires significant human annotation. Although deep neural network approaches, such as DNSMOS and UTMOS, have been developed to predict MOS to avoid this issue, they often suffer from insufficient training data. Recognizing that the comparison of speech enhancement (SE) systems prioritizes a reliable system comparison over absolute scores, we propose URGENT-PK, a novel ranking approach leveraging pairwise comparisons. URGENT-PK takes homologous enhanced speech pairs as input to predict relative quality rankings. This pairwise paradigm efficiently utilizes limited training data, as all pairwise permutations of multiple systems constitute a training instance. Experiments across multiple open test sets demonstrate URGENT-PK's superior system-level ranking performance over state-of-the-art baselines, despite its simple network architecture and limited training data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_23874 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | URGENT-PK: Perceptually-Aligned Ranking Model Designed for Speech Enhancement Competition Wang, Jiahe Li, Chenda Wang, Wei Zhang, Wangyou Cornell, Samuele Sach, Marvin Scheibler, Robin Saijo, Kohei Fu, Yihui Ni, Zhaoheng Kumar, Anurag Fingscheidt, Tim Watanabe, Shinji Qian, Yanmin Audio and Speech Processing Sound The Mean Opinion Score (MOS) is fundamental to speech quality assessment. However, its acquisition requires significant human annotation. Although deep neural network approaches, such as DNSMOS and UTMOS, have been developed to predict MOS to avoid this issue, they often suffer from insufficient training data. Recognizing that the comparison of speech enhancement (SE) systems prioritizes a reliable system comparison over absolute scores, we propose URGENT-PK, a novel ranking approach leveraging pairwise comparisons. URGENT-PK takes homologous enhanced speech pairs as input to predict relative quality rankings. This pairwise paradigm efficiently utilizes limited training data, as all pairwise permutations of multiple systems constitute a training instance. Experiments across multiple open test sets demonstrate URGENT-PK's superior system-level ranking performance over state-of-the-art baselines, despite its simple network architecture and limited training data. |
| title | URGENT-PK: Perceptually-Aligned Ranking Model Designed for Speech Enhancement Competition |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2506.23874 |