Prediction-adaptation-correction recurrent neural networks for low-resource language speech recognition

Yu Zhang; Ekapol Chuangsuwanich; James R. Glass; Dong Yu

ICASSP 2016accepted0 citations

Prediction-adaptation-correction recurrent neural networks for low-resource language speech recognition

Yu Zhang, Ekapol Chuangsuwanich, James R. Glass, Dong Yu

Abstract

In this paper, we investigate the use of prediction-adaptation-correction recurrent neural networks (PAC-RNNs) for low-resource speech recognition. A PAC-RNN is comprised of a pair of neural networks in which a correction network uses auxiliary information given by a prediction network to help estimate the state probability. The information from the correction network is also used by the prediction network in a recurrent loop. Our model outperforms other state-of-the-art neural networks (DNNs, LSTMs) on IARPA-Babel tasks. Moreover, transfer learning from a language that is similar to the target language can help improve performance further.

BibTeX

@inproceedings{icassp2016_predictionadapta,
  title = {Prediction-adaptation-correction recurrent neural networks for low-resource language speech recognition},
  author = {Yu Zhang and Ekapol Chuangsuwanich and James R. Glass and Dong Yu},
  booktitle = {ICASSP 2016},
  year = {2016}
}