Name
Abstract | ||
|---|---|---|
Linear Discriminant Analysis (LDA) is a feature selection method in speech recognition. LDA finds transformations that maximizes the between-class scatter and minimizes within-class scatter. This transformation can be obtained in a class-dependent or class independent manner. In this paper, we propose a method to improve LDA and also we use it instead of DCT in MFCC extraction. This transformation matrix is computed through three evolutionary methods (GA, HS, and PSO) to optimize class-dependent LDA transformation matrix for robust MFCC extraction. For this purpose, we first use logarithm of clean speech Mel filter bank energies (LMFE) of each class to define within-class scatter for each class and between-class scatter for over all classes. Next, class-dependent transformation matrix is utilized in place of DCT in MFCC feature extraction. The experimental results show that the proposed speech recognition and optimization methods using class-dependent LDA, achieves a significant isolated word recognition rate on Aurora2 database. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1145/1570256.1570285 | GECCO (Companion) |
Keywords | Field | DocType |
evolutionary class-dependent,class-dependent lda,mfcc feature extraction,within-class scatter,minimizes within-class scatter,class-dependent transformation matrix,class independent manner,between-class scatter,robust speech recognition,mfcc extraction,transformation matrix,class-dependent lda transformation matrix,filter bank,particle swarm optimization,speech recognition,feature extraction,mfcc,feature selection,word recognition,harmony search | Mel-frequency cepstrum,Feature selection,Computer science,Filter bank,Discrete cosine transform,Artificial intelligence,Logarithm,Transformation matrix,Pattern recognition,Feature extraction,Speech recognition,Linear discriminant analysis,Machine learning | Conference |
Citations | PageRank | References |
2 | 0.40 | 9 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hossein Moeinzadeh | 1 | 23 | 4.07 |
| Mehdi Mohammadi | 2 | 1091 | 50.02 |
| Ahmad Akbari | 3 | 159 | 23.17 |
| Babak Nasersharif | 4 | 88 | 13.21 |