Abstract:
Sequence-to-sequence models have shown success in end-to-end speech recognition. However these models have only used shallow acoustic encoder networks. In our work, we su...Show MoreMetadata
Abstract:
Sequence-to-sequence models have shown success in end-to-end speech recognition. However these models have only used shallow acoustic encoder networks. In our work, we successively train very deep convolutional networks to add more expressive power and better generalization for end-to-end ASR models. We apply network-in-network principles, batch normalization, residual connections and convolutional LSTMs to build very deep recurrent and convolutional structures. Our models exploit the spectral structure in the feature space and add computational depth without overfitting issues. We experiment with the WSJ ASR task and achieve 10.5% word error rate without any dictionary or language model using a 15 layer deep network.
Published in: 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Date of Conference: 05-09 March 2017
Date Added to IEEE Xplore: 19 June 2017
ISBN Information:
Electronic ISSN: 2379-190X