Deep recurrent de-noising auto-encoder and blind de-reverberation for reverberated speech recognition
Acoustics, Speech and Signal Processing(2014)
摘要
This paper describes our joint efforts to provide robust automatic speech recognition (ASR) for reverberated environments, such as in hands-free human-machine interaction. We investigate blind feature space de-reverberation and deep recurrent de-noising auto-encoders (DAE) in an early fusion scheme. Results on the 2014 REVERB Challenge development set indicate that the DAE front-end provides complementary performance gains to multi-condition training, feature transformations, and model adaptation. The proposed ASR system achieves word error rates of 17.62 % and 36.6 % on simulated and real data, which is a significant improvement over the Challenge baseline (25.16 and 47.2 %).
更多查看译文
关键词
feature extraction,recurrent neural nets,reverberation,signal denoising,speech codecs,speech recognition,ASR,automatic speech recognition,blind dereverberation,blind feature space dereverberation,deep recurrent denoising auto-encoder,feature transformations,fusion scheme,hands-free human-machine interaction,model adaptation,multi-condition training,reverberated environments,reverberated speech recognition,word error rates,De-reverberation,automatic speech recognition,feature enhancement,recurrent neural networks
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要