arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1803.02551cs.CLcs.LGcs.SDeess.ASstat.ML
收藏

基于无监督学习的鲁棒自动语音识别域不变特征提取

Extracting Domain Invariant Features by Unsupervised Learning for Robust Automatic Speech Recognition

MIT Computer Science and Artificial Intelligence Laboratory(麻省理工学院计算机科学与人工智能实验室)

Wei-Ning Hsu, James Glass

摘要:

自动语音识别(ASR)系统的性能可能会受到以前看不见的条件的严重影响,这通常是由于训练和测试分布之间的不匹配。在本文中,我们通过研究域不变特征来解决鲁棒性问题,使域信息对ASR系统透明,解决不匹配问题。具体来说,我们研究了最近的模型,称为因子化分层变分自编码器(FHVAE)。FHVAE学习在没有监督的情况下将序列级和段级属性分解为不同的潜变量。我们认为,包含段级信息的一组潜变量是我们所需的域不变的ASR功能。在Aurora-4和CHiME-4上进行了实验,在不匹配的域上分别显示出41%和27%的绝对字出错率降低。

英文摘要:

The performance of automatic speech recognition (ASR) systems can be significantly compromised by previously unseen conditions, which is typically due to a mismatch between training and testing distributions. In this paper, we address robustness by studying domain invariant features, such that domain information becomes transparent to ASR systems, resolving the mismatch problem. Specifically, we investigate a recent model, called the Factorized Hierarchical Variational Autoencoder (FHVAE). FHVAEs learn to factorize sequence-level and segment-level attributes into different latent variables without supervision. We argue that the set of latent variables that contain segment-level information is our desired domain invariant feature for ASR. Experiments are conducted on Aurora-4 and CHiME-4, which demonstrate 41% and 27% absolute word error rate reductions respectively on mismatched domains.

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。

↑