MADI：跨域语音识别的域间匹配和域内判别
MADI: Inter-domain Matching and Intra-domain Discrimination for Cross-domain Speech Recognition
摘要：
端到端自动语音识别（ASR）在应用到新的领域时，通常会由于领域转换而导致性能下降。无监督领域自适应（UDA）通过将知识从源领域转移到目标领域来提高未标记目标领域的性能。为了提高模型的可移植性，现有的UDA方法主要集中在全局和/或局部地匹配源域和目标域的分布，而忽略了模型的可区分性。本文提出了一种新的基于域间匹配和域内区分（MADI）的ASR UDA方法，该方法通过细粒度的域间匹配提高模型的可移植性，同时通过域内对比判别提高模型的可区分性。Libri-Adapt数据集上的评估证明了我们方法的有效性。MADI将跨设备和跨环境ASR的相对字出错率（WER）分别降低了17.7%和22.8%。
英文摘要：
End-to-end automatic speech recognition (ASR) usually suffers from performance degradation when applied to a new domain due to domain shift. Unsupervised domain adaptation (UDA) aims to improve the performance on the unlabeled target domain by transferring knowledge from the source to the target domain. To improve transferability, existing UDA approaches mainly focus on matching the distributions of the source and target domains globally and/or locally, while ignoring the model discriminability. In this paper, we propose a novel UDA approach for ASR via inter-domain MAtching and intra-domain DIscrimination (MADI), which improves the model transferability by fine-grained inter-domain matching and discriminability by intra-domain contrastive discrimination simultaneously. Evaluations on the Libri-Adapt dataset demonstrate the effectiveness of our approach. MADI reduces the relative word error rate (WER) on cross-device and cross-environment ASR by 17.7% and 22.8%, respectively.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
