arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1608.00892cs.CL
收藏

小足迹公路网的知识蒸馏

Knowledge Distillation for Small-footprint Highway Networks

The University of Edinburgh(爱丁堡大学)

Liang Lu, Michelle Guo, Steve Renals

摘要:

在过去的几年里,深度学习极大地推进了语音识别的发展。然而,与传统的高斯混合声学模型相比,神经网络模型通常要大得多,并且因此在嵌入式设备中不是非常可部署的。在此之前,我们研究了一种用于声学建模的紧凑型高速公路深度神经网络(HDNN),它是一种深度门控前馈神经网络。我们已经表明,基于HDNN的声学模型可以实现相当的识别精度与更少的模型参数相比,平原深神经网络(DNN)的声学模型。在本文中,我们通过利用知识蒸馏技术进一步推动边界,该技术也被称为{ \it师生}培训,即,在高精度冗余模型的监督下,对紧凑型HDNN模型进行训练。此外,我们还探讨了师生培训背景下的序列培训和适应。我们的实验是在AMI会议语音识别语料库上进行的。通过这项技术,我们显著提高了HDNN声学模型的识别精度,参数不到80万,缩小了该模型与3000万参数的普通DNN之间的差距。

英文摘要:

Deep learning has significantly advanced state-of-the-art of speech recognition in the past few years. However, compared to conventional Gaussian mixture acoustic models, neural network models are usually much larger, and are therefore not very deployable in embedded devices. Previously, we investigated a compact highway deep neural network (HDNN) for acoustic modelling, which is a type of depth-gated feedforward neural network. We have shown that HDNN-based acoustic models can achieve comparable recognition accuracy with much smaller number of model parameters compared to plain deep neural network (DNN) acoustic models. In this paper, we push the boundary further by leveraging on the knowledge distillation technique that is also known as {\it teacher-student} training, i.e., we train the compact HDNN model with the supervision of a high accuracy cumbersome model. Furthermore, we also investigate sequence training and adaptation in the context of teacher-student training. Our experiments were performed on the AMI meeting speech recognition corpus. With this technique, we significantly improved the recognition accuracy of the HDNN acoustic model with less than 0.8 million parameters, and narrowed the gap between this model and the plain DNN with 30 million parameters.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑