Abstract | ||
---|---|---|
This paper presents comparison and optimization of acoustic features for source cell-phone recognition using recorded speech signals. Different acoustic feature extraction methods such as Mel-frequency, linear frequency and Bark frequency cepstral coefficients (MFCC, LFCC and BFCC) and linear prediction cepstral coefficients (LPCC) are considered. In addition to different feature sets, the effect of dynamic features, delta and double-delta coefficients (Δ and Δ2), and feature normalizations, cepstral mean normalization (CMN), cepstral variance normalization (CVN) and cepstral mean and variance normalization (CMVN) are also examined on the performance of source cell-phone recognition. The same support vector machine (SVM) classifier with fixed parameters and the same cell-phone dataset are used in the experiments in order to make a fair comparison of different features and feature normalization techniques. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1145/2482513.2482520 | IH&MMSec |
Keywords | Field | DocType |
source cell-phone recognition,dynamic feature,speech signal,cepstral mean,bark frequency cepstral coefficient,different acoustic feature extraction,linear prediction cepstral coefficient,different feature,different feature set,acoustic feature,cepstral variance normalization | Audio forensics,Mel-frequency cepstrum,Normalization (statistics),Pattern recognition,Computer science,Cepstrum,Support vector machine,Feature extraction,Speech recognition,Cepstral Mean and Variance Normalization,Artificial intelligence,Classifier (linguistics) | Conference |
Citations | PageRank | References |
3 | 0.40 | 24 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Cemal Hanilçi | 1 | 171 | 11.23 |
Figen Ertas | 2 | 40 | 2.92 |