Research on Speech Recognition Based on Improved LSTM Deep Neural Network
[1]Zhao Shufang,Dong Xiaoyu.Research on Speech Recognition Based on Improved LSTM Deep Neural Network[J].Journal of Zhengzhou University (Engineering Science),2018,39(05):63-67.[doi:10.13705/j.issn.1671-6833.2018.02.004]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
39
Number of periods:
2018 05
Page number:
63-67
Column:
Public date:
2018-08-21
- Title:
- Research on Speech Recognition Based on Improved LSTM Deep Neural Network
- CLC:
- -
- DOI:
- 10.13705/j.issn.1671-6833.2018.02.004
- Abstract:
-
The language model based on neural network LSTM structure, the LSTM structure used in the hidden layer unit, the structure unit comprises a memory unit which can store the information for a long time, which has a good memory function for the historical information. But the LSTM in the current input information state9 does not affect the final output information of the output gate, get less historical information. To solve the above problems, this paper puts forward based on improved LSTM (long short-term memory) modeling method of network model. The model increases the connection from the current input gate to the output gate, and simultaneously combines the oblivious gate and the input gate into a single update. The door keeper input and forgotten past and present memory consolidation, can choose to forget before the accumulation of information, the improved LSTM model can learn the long history of information, solve the drawback of the LSTM method is morerobust. This paper uses the neural network languag LSTM model based on the inproved model on TIMIT data sets show that the axxuracy of test. The results illustrate that the improved LSTM identification error rate is 5
% lower than the standard LSTM identification error rate.
- References:
- -
- Similar References:
Memo
-
Last Update:
2018-08-22