Back to Search
Start Over
Cross-utterance ASR Rescoring with Graph-based Label Propagation
- Source :
- Proc. IEEE ICASSP, June 2023
- Publication Year :
- 2023
-
Abstract
- We propose a novel approach for ASR N-best hypothesis rescoring with graph-based label propagation by leveraging cross-utterance acoustic similarity. In contrast to conventional neural language model (LM) based ASR rescoring/reranking models, our approach focuses on acoustic information and conducts the rescoring collaboratively among utterances, instead of individually. Experiments on the VCTK dataset demonstrate that our approach consistently improves ASR performance, as well as fairness across speaker groups with different accents. Our approach provides a low-cost solution for mitigating the majoritarian bias of ASR systems, without the need to train new domain- or accent-specific models.<br />Comment: To appear in IEEE ICASSP 2023
Details
- Database :
- arXiv
- Journal :
- Proc. IEEE ICASSP, June 2023
- Publication Type :
- Report
- Accession number :
- edsarx.2303.15132
- Document Type :
- Working Paper
- Full Text :
- https://doi.org/10.1109/ICASSP49357.2023.10096820