Limits...
Using hierarchical time series clustering algorithm and wavelet classifier for biometric voice classification.

Fong S - J. Biomed. Biotechnol. (2012)

Bottom Line: Lately voice classification is found useful in phone monitoring, classifying speakers' gender, ethnicity and emotion states, and so forth.In this paper, a collection of computational algorithms are proposed to support voice classification; the algorithms are a combination of hierarchical clustering, dynamic time wrap transform, discrete wavelet transform, and decision tree.The proposed algorithms are relatively more transparent and interpretable than the existing ones, though many techniques such as Artificial Neural Networks, Support Vector Machine, and Hidden Markov Model (which inherently function like a black box) have been applied for voice verification and voice identification.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Science, University of Macau, Taipa, Macau. ccfong@umac.mo

ABSTRACT
Voice biometrics has a long history in biosecurity applications such as verification and identification based on characteristics of the human voice. The other application called voice classification which has its important role in grouping unlabelled voice samples, however, has not been widely studied in research. Lately voice classification is found useful in phone monitoring, classifying speakers' gender, ethnicity and emotion states, and so forth. In this paper, a collection of computational algorithms are proposed to support voice classification; the algorithms are a combination of hierarchical clustering, dynamic time wrap transform, discrete wavelet transform, and decision tree. The proposed algorithms are relatively more transparent and interpretable than the existing ones, though many techniques such as Artificial Neural Networks, Support Vector Machine, and Hidden Markov Model (which inherently function like a black box) have been applied for voice verification and voice identification. Two datasets, one that is generated synthetically and the other one empirically collected from past voice recognition experiment, are used to verify and demonstrate the effectiveness of our proposed voice classification algorithm.

Show MeSH

Related in: MedlinePlus

Snapshots of a decision tree as a result of building a classifier by (a) using the original time series, and (b) using the transformed wavelets.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3351073&req=5

fig10: Snapshots of a decision tree as a result of building a classifier by (a) using the original time series, and (b) using the transformed wavelets.

Mentions: The performance comparison table is given in Table 3. It compares mainly the classification accuracy by using a J48 decision tree in WEKA of the time series version and the Wavelet version of the two testing datasets. It can be noticed that in general Wavelets have improvement over the time series in terms of classification accuracy. The results of the empirical data are generally lower in accuracy than the synthetic control data probably due to its complex and less uniform in the time series structures, plus the normalization effect for limiting the time series into fixed length from its original variable length. However, wavelet transformation still shows its advantage in applying to the empirical data. A sample of the decision tree generated from the experiment is shown in Figure 10. By using the decision tree as classifier, new voiceprint can fit into a specific class by traversing the decision tree.


Using hierarchical time series clustering algorithm and wavelet classifier for biometric voice classification.

Fong S - J. Biomed. Biotechnol. (2012)

Snapshots of a decision tree as a result of building a classifier by (a) using the original time series, and (b) using the transformed wavelets.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3351073&req=5

fig10: Snapshots of a decision tree as a result of building a classifier by (a) using the original time series, and (b) using the transformed wavelets.
Mentions: The performance comparison table is given in Table 3. It compares mainly the classification accuracy by using a J48 decision tree in WEKA of the time series version and the Wavelet version of the two testing datasets. It can be noticed that in general Wavelets have improvement over the time series in terms of classification accuracy. The results of the empirical data are generally lower in accuracy than the synthetic control data probably due to its complex and less uniform in the time series structures, plus the normalization effect for limiting the time series into fixed length from its original variable length. However, wavelet transformation still shows its advantage in applying to the empirical data. A sample of the decision tree generated from the experiment is shown in Figure 10. By using the decision tree as classifier, new voiceprint can fit into a specific class by traversing the decision tree.

Bottom Line: Lately voice classification is found useful in phone monitoring, classifying speakers' gender, ethnicity and emotion states, and so forth.In this paper, a collection of computational algorithms are proposed to support voice classification; the algorithms are a combination of hierarchical clustering, dynamic time wrap transform, discrete wavelet transform, and decision tree.The proposed algorithms are relatively more transparent and interpretable than the existing ones, though many techniques such as Artificial Neural Networks, Support Vector Machine, and Hidden Markov Model (which inherently function like a black box) have been applied for voice verification and voice identification.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Science, University of Macau, Taipa, Macau. ccfong@umac.mo

ABSTRACT
Voice biometrics has a long history in biosecurity applications such as verification and identification based on characteristics of the human voice. The other application called voice classification which has its important role in grouping unlabelled voice samples, however, has not been widely studied in research. Lately voice classification is found useful in phone monitoring, classifying speakers' gender, ethnicity and emotion states, and so forth. In this paper, a collection of computational algorithms are proposed to support voice classification; the algorithms are a combination of hierarchical clustering, dynamic time wrap transform, discrete wavelet transform, and decision tree. The proposed algorithms are relatively more transparent and interpretable than the existing ones, though many techniques such as Artificial Neural Networks, Support Vector Machine, and Hidden Markov Model (which inherently function like a black box) have been applied for voice verification and voice identification. Two datasets, one that is generated synthetically and the other one empirically collected from past voice recognition experiment, are used to verify and demonstrate the effectiveness of our proposed voice classification algorithm.

Show MeSH
Related in: MedlinePlus