基于改进的长短期记忆神经网络方言辨识模型  

Dialect Identification Model Based on Improved Long Short-term Memory Neural Network

作　　者：艾虎 李菲 AI Hu;LI Fei(Department of Criminal Technology,Guizhou Police College,Guiyang 550005,China;Faculty of Humanities,The Education University of Hong Kong,Hong Kong 999077,China)

机构地区：[1]贵州警察学院刑事技术系,贵阳550005 [2]香港教育大学人文学院,香港999077

出　　处：《科学技术与工程》2019年第2期163-169,共7页Science Technology and Engineering

基　　金：贵州省科技计划(黔科合[2016]支撑2847)资助

摘　　要：在案件侦破中,方言的辨别能提供重要线索。为了对汉语方言进行辨别,基于长短期记忆神经网络(LSTM)的方言辨识模型被提出,语音样本数据,其中包括地区口头禅,均采集于贵州省6个地区,并提取梅尔频率倒谱系数(MFCC),每份语音样本MFCC后面加上相应的地区口头禅MFCC,然后采用滑窗进行信息重叠分块,对每块分别进行横向与纵向奇异值分解并保留高贡献率的特征向量,把分块合并作为方言辨识模型的输入数据。先对LSTM进行改进,然后构建方言辨识模型。通过交叉实验对该模型进行训练和验证,从而对滑窗的宽度进行优化,同时与循环神经网络(RNN)进行比较。实验结果证明研究构建的LSTM模型对汉语方言辨识是高效的。Chinese dialect identifications may provide important clues for forensic investigation.An effective dialect identification model has keen proposed for Chinese dialect identification based on improved long short-term memory(LSTM).Mel frequency cepstral coefficients(MFCC)was extracted from speech samples including regional pet phrase collected from six regions in Guizhou province,then added a corresponding regional pet phrase after each voice sample,and then used the sliding window to conduct information overlapping blocking.The singular value of each block was decomposed from horizontal and vertical and high contribution rate feature vectors were retained,and the blocks were combined as the input data of the dialect identification model.Firstly,the LSTM is improved,then a dialect identification model is constructed,and the model is trained and verified by adopting an experiment,so that the width of the sliding window are optimized and the LSTM is compared with recurrent neural network(RNN).The experimental results show that the model based on improved LSTM is efficient for Chinese dialect identification.

关 键 词：汉语方言辨识 梅尔频率倒谱系数 地区口头禅 奇异值分解 长短期记忆神经网络 

分 类 号：TP391.42[自动化与计算机技术—计算机应用技术]