Cross-utterance ASR Rescoring with Graph-based Label Propagation

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.

To appear in IEEE ICASSP 2023

References in corpus (6)

• Sequence Transduction with Recurrent Neural Networks
• Toward Fairness in Speech Recognition: Discovery and mitigation of performance disparities
• Improving fairness in speaker verification via Group-adapted Fusion Network
• Deep Graph Random Process for Relational-Thinking-Based Speech Recognition
• Cross-utterance Reranking Models with BERT and Graph Convolutional Networks for Conversational Speech Recognition
• Graph-based Multi-View Fusion and Local Adaptation: Mitigating Within-Household Confusability for Speaker Identification