IMPROVING THE ROBUSTNESS OF AUTOMATIC SPEECH RECOGNITION

机译：提高自动语音识别的鲁棒性

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Automatic speech recognition systems present performance that degrades dramatically in adverse situations, in the presence of noise or with different speakers. These degradations are due to the differences that occur between training and testing conditions. Speech recognition in adverse conditions has received increased attention during the last decade, since noise resistance has become one of the major bottlenecks for practical use of speech recognizers in real life. This paper reviews some of the methods proposed so far at the various stages of the recognition process in order to improve the robustness of the systems in real life conditions.

机译：自动语音识别系统存在在噪声或不同扬声器的情况下在不利情况下显着降低的性能。这些降解是由于训练和测试条件之间发生的差异。在过去十年中，不利条件的语音识别受到了更多的关注，因为抗噪声已成为现实生活中的语音识别员实际应用的主要瓶颈之一。本文审查了迄今为止识别过程的各个阶段提出的一些方法，以改善现实生活条件下系统的稳健性。

著录项

来源
《Conference on Speech Technology and Human-Computer Dialogue》|2007年||共8页
会议地点
作者
Jean-Paul HATON;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类 TN912-53;
关键词
Automatic speech recognition; Robustness; Speech enhancement; Parameterization; Model adaptation; Missing data;

机译：自动语音识别;鲁棒性;语音增强;参数化;模型适应;缺少数据;

相似文献

外文文献
中文文献
专利

1. Unsupervised Speech Enhancement Based on Multichannel NMF-Informed Beamforming for Noise-Robust Automatic Speech Recognition [J] . Shimada Kazuki, Bando Yoshiaki, Mimura Masato, Audio, Speech, and Language Processing, IEEE/ACM Transactions on . 2019,第5期

机译：基于多通道NMF信息波束形成的无监督语音增强技术，用于强噪声自动语音识别
2. Unsupervised Speech Enhancement Based on Multichannel NMF-Informed Beamforming for Noise-Robust Automatic Speech Recognition [J] . Shimada Kazuki, Bando Yoshiaki, Mimura Masato, Audio, Speech, and Language Processing, IEEE/ACM Transactions on . 2019,第5期

机译：基于多通道NMF的噪声强度自动语音识别的无监督语音增强
3. Multi-Channel Speech Enhancement and Amplitude Modulation Analysis for Noise Robust Automatic Speech Recognition [J] . Moritz Niko, Adiloǧlu Kamil, Anemüller Jörn, Computer speech and language . 2017,第nova期

机译：噪声鲁棒自动语音识别的多通道语音增强和幅度调制分析
4. A study on bias-based speech signal conditioning techniques for improving the robustness of automatic speech recognition [C] . Chowdhury M., Selouani S.-A., OShaughnessy D. Electrical and Computer Engineering, 2009. CCECE '09 . 2009

机译：基于偏置的语音信号调理技术研究，旨在提高自动语音识别的鲁棒性
5. Compressive nonlinearity for representing speech spectral magnitude to improve noise robustness of automatic speech recognition . [D] . Wong, Brian. 2011

机译：压缩非线性表示语音频谱幅度提高语音自动识别的鲁棒性。
6. Brain-inspired speech segmentation for automatic speech recognition using the speech envelope as a temporal reference [O] . Byeongwook Lee, Kwang-Hyun Cho -1

机译：以语音包络作为时间参考的自动语音识别的大脑启发式语音分割
7. Improving Noise Robust Automatic Speech Recognition with Single-Channel Time-Domain Enhancement Network [O] . Keisuke Kinoshita, Tsubasa Ochiai, Marc Delcroix, 2020

机译：用单通道时域增强网络提高噪声鲁棒自动语音识别

IMPROVING THE ROBUSTNESS OF AUTOMATIC SPEECH RECOGNITION

摘要

著录项

相似文献

相关主题

期刊订阅