In order to improve the performance for far-field speech recognition, this paper proposes to distill knowledge from the close-talking model to the far-field model using parallel data. The close-talking model is called the teacher model. The far-field model is called the student model. The student model is trained to imitate the output distributions of the teacher model. This constraint can be realized by minimizing the Kullback-Leibler (KL) divergence between the output distribution of the student model and the teacher model. Experimental results on AMI corpus show that the best student model achieves up to 4.7% absolute word error rate (WER) reduction when compared with the conventionally-trained baseline models.
@article{arxiv.1802.06941,
title = {Distilling Knowledge Using Parallel Data for Far-field Speech Recognition},
author = {Jiangyan Yi and Jianhua Tao and Zhengqi Wen and Bin Liu},
journal= {arXiv preprint arXiv:1802.06941},
year = {2018}
}