“Askar - 130520”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
第8行: 第8行:
  
 
1. train models basend on MFCC parameters(MFCC_0_D_A) and compare with the previous experiment results.
 
1. train models basend on MFCC parameters(MFCC_0_D_A) and compare with the previous experiment results.
 +
 
2. finish the reasing of the tutorial.
 
2. finish the reasing of the tutorial.

2013年5月20日 (一) 04:57的最后版本

works finished:

1. extracted MFCC parameters(MFCC_0_D_A) and working on training UBM.

2. almost finished reading tutorial--A Gentle Tutorial of the EM Algorithm and its Application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models.

work plan:

1. train models basend on MFCC parameters(MFCC_0_D_A) and compare with the previous experiment results.

2. finish the reasing of the tutorial.