Hunan Dialects Identification Based on GRU-HMM Acoustic Model
XIE Kexin
DONG Hu
ZOU Xiao
TANG Chen
QIAN Shengyou
Abstract：An acoustic model based on Gated Recurrent Unit(GRU)neural networks and Hidden Markov Model(HMM)is established. The Mel-Frequency Cepstral Coefficients(MFCC)is used as the input of the acoustic model,and the GRU neural net?work can be used to perform the probability statistics on the speech data in real time. The obtained probability values are statistically re-evaluated by the HMM model. Finally the identification results are obtained. This method is used to identify the Hunan dialect. Experiments show that this acoustic model has better identification efficiency than the traditional acoustic model.
Keywords：gated recurrent unit(GRU)hidden markov Model(HMM)acoustic modelmel-frequency cepstral coeffi?cients(MFCC)Hunan dialect identification
Publication Date：2019-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：4( 493-496 )