面向包容与公平语音识别的重口音语音聚类与挖掘
Clustering and Mining Accented Speech for Inclusive and Fair Speech Recognition
AI总结：
本文提出口音聚类与挖掘方案，结合预训练、分布鲁棒优化和无监督聚类，在挖掘出的印度口音语音上微调ASR，相对随机采样分别提升10.0%和5.3%，实现更公平的语音识别。
摘要：
现代自动语音识别（ASR）系统通常使用超过数万小时的语音数据进行训练，这是其取得巨大成功的主要因素之一。然而，这类数据的分布通常偏向常见口音或典型语音模式，导致这些系统在非典型重口音语音上往往表现不佳。本文提出了面向公平语音识别系统的口音聚类与挖掘方案，使其在代表性不足的重口音语音上也能表现同样出色。针对口音识别，我们采用了三种方案来克服有监督口音数据规模有限的问题：有监督或无监督预训练、分布鲁棒优化（DRO）以及无监督聚类。这三种方案能够显著提升口音识别模型性能，尤其是在不平衡和小规模的重口音语音场景下。使用所提出的有监督或无监督聚类方案，在挖掘出的印度口音语音上微调ASR，与在随机采样语音上微调相比，分别取得了10.0%和5.3%的相对改进。
英文摘要：
Modern automatic speech recognition (ASR) systems are typically trained on more than tens of thousands hours of speech data, which is one of the main factors for their great success. However, the distribution of such data is typically biased towards common accents or typical speech patterns. As a result, those systems often poorly perform on atypical accented speech. In this paper, we present accent clustering and mining schemes for fair speech recognition systems which can perform equally well on under-represented accented speech. For accent recognition, we applied three schemes to overcome limited size of supervised accent data: supervised or unsupervised pre-training, distributionally robust optimization (DRO) and unsupervised clustering. Three schemes can significantly improve the accent recognition model especially for unbalanced and small accented speech. Fine-tuning ASR on the mined Indian accent speech using the proposed supervised or unsupervised clustering schemes showed 10.0% and 5.3% relative improvements compared to fine-tuning on the randomly sampled speech, 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。
