期刊
SPEECH COMMUNICATION
卷 89, 期 -, 页码 103-112出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.specom.2017.03.003
关键词
Automatic speech recognition; Articulatory trajectories; Vocal tract variables; Hybrid convolutional neural networks; Time-frequency convolution; Convolutional neural networks
资金
- NSF [IIS-0964556, IIS-1162046, IIS-1161962]
- Direct For Computer & Info Scie & Enginr
- Div Of Information & Intelligent Systems [1162033] Funding Source: National Science Foundation
Studies have shown that articulatory information helps model speech variability and, consequently, improves speech recognition performance. But learning speaker-invariant articulatory models is challenging, as speaker-specific signatures in both the articulatory and acoustic space increase complexity of speech-to-articulatory mapping, which is already an ill-posed problem due to its inherent nonlinearity and non unique nature. This work explores using deep neural networks (DNNs) and convolutional neural networks (CNNs) for mapping speech data into its corresponding articulatory space. Our speech-inversion results indicate that the CNN models perform better than their DNN counterparts. In addition, we use these inverse-models to generate articulatory information from speech for two separate speech recognition tasks: the WSJ1 and Aurora-4 continuous speech recognition tasks. This work proposes a hybrid convolutional neural network (HCNN), where two parallel layers are used to jointly model the acoustic and articulatory spaces, and the decisions from the parallel layers are fused at the output context-dependent (CD) state level. The acoustic model performs time-frequency convolution on filterbank-energy-level features, whereas the articulatory model performs time convolution on the articulatory features. The performance of the proposed architecture is compared to that of the CNN- and DNN-based systems using gammatone filterbank energies as acoustic features, and the results indicate that the HCNN-based model demonstrates lower word error rates compared to the CNN/DNN baseline systems. (C) 2017 Elsevier B.V. All rights reserved.
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