Title | ||
---|---|---|
Elastic Spectral Distortion For Low Resource Speech Recognition With Deep Neural Networks |
Abstract | ||
---|---|---|
An acoustic model based on hidden Markov models with deep neural networks (DNN-HMM) has recently been proposed and achieved high recognition accuracy. In this paper, we investigated an elastic spectral distortion method to artificially augment training samples to help DNN-HMMs acquire enough robustness even when there are a limited number of training samples. We investigated three distortion methods-vocal tract length distortion, speech rate distortion, and frequency-axis random distortion and evaluated those methods with Japanese lecture recordings. In a large vocabulary continuous speech recognition task with only 10 hours of training samples, a DNN-HMM trained with the elastic spectral distortion method achieved a 10.1% relative word error reduction compared with a normally trained DNN-HMM. |
Year | Venue | Keywords |
---|---|---|
2013 | 2013 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING (ASRU) | Deep neural network, speech recognition, elastic distortion |
DocType | Citations | PageRank |
Conference | 12 | 0.86 |
References | Authors | |
8 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Naoyuki Kanda | 1 | 103 | 19.45 |
Ryu Takeda | 2 | 66 | 11.25 |
Yasunari Obuchi | 3 | 64 | 12.71 |