arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.00560cs.LGcs.CLstat.ML
收藏

关键词识别中的权重重要性稀疏训练

Weight-importance sparse training in keyword spotting

NIO Co., Ltd(蔚来汽车有限公司)

Sihao Xue, Zhenyi Ying, Fan Mo, Min Wang, Jue Sun

摘要:

近年来,为了处理复杂的语音识别问题,语音识别系统采用了大规模的语音模型。这些模型中的参数数量使得它们难以部署,特别是在一些资源短缺的设备上,如车载平板电脑。此外,在大多数情况下,ASR系统用于处理实时问题,如关键字定位(KWS)。这与大型模型需要较长计算时间的事实相矛盾。为了解决这个问题,我们应用一些稀疏算法来减少一些广泛使用的模型,深度神经网络(DNN)KWS中的参数数量,这需要非常短的计算时间。模型中的参数可以在极小的影响衰减下被删减90%甚至95%以上.稀疏模型的性能优于参数阶数相同的基线模型。此外,稀疏算法可以使我们自动地找到合理的模型大小,而无需考虑选择原始模型大小。

英文摘要:

Large size models are implemented in recently ASR system to deal with complex speech recognition problems. The num- ber of parameters in these models makes them hard to deploy, especially on some resource-short devices such as car tablet. Besides this, at most of time, ASR system is used to deal with real-time problem such as keyword spotting (KWS). It is contradictory to the fact that large model requires long com- putation time. To deal with this problem, we apply some sparse algo- rithms to reduces number of parameters in some widely used models, Deep Neural Network (DNN) KWS, which requires real short computation time. We can prune more than 90 % even 95% of parameters in the model with tiny effect decline. And the sparse model performs better than baseline models which has same order number of parameters. Besides this, sparse algorithm can lead us to find rational model size au- tomatically for certain problem without concerning choosing an original model size.

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。

↑