期刊
EURASIP JOURNAL ON AUDIO SPEECH AND MUSIC PROCESSING
卷 -, 期 -, 页码 -出版社
SPRINGER
DOI: 10.1186/s13636-014-0047-0
关键词
Speech recognition; Deep neural network; Noise injection
资金
- National Science Foundation of China (NSFC) [61371136, 61271389]
- National Basic Research Program (973 Program) of China [2013CB329302]
- MESTDC PhD Foundation [20130002120011]
- Sinovoice and Huilan Ltd.
Deep neural networks (DNNs) have gained remarkable success in speech recognition, partially attributed to the flexibility of DNN models in learning complex patterns of speech signals. This flexibility, however, may lead to serious over-fitting and hence miserable performance degradation in adverse acoustic conditions such as those with high ambient noises. We propose a noisy training approach to tackle this problem: by injecting moderate noises into the training data intentionally and randomly, more generalizable DNN models can be learned. This 'noise injection' technique, although known to the neural computation community already, has not been studied with DNNs which involve a highly complex objective function. The experiments presented in this paper confirm that the noisy training approach works well for the DNN model and can provide substantial performance improvement for DNN-based speech recognition.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据