Wavelet-Based Mel-Frequency Cepstral Coefficients for Speaker Identification using Hidden Markov Models - Computer Science > SoundReport as inadecuate




Wavelet-Based Mel-Frequency Cepstral Coefficients for Speaker Identification using Hidden Markov Models - Computer Science > Sound - Download this document for free, or read online. Document in PDF available to download.

Abstract: To improve the performance of speaker identification systems, an effectiveand robust method is proposed to extract speech features, capable of operatingin noisy environment. Based on the time-frequency multi-resolution property ofwavelet transform, the input speech signal is decomposed into various frequencychannels. For capturing the characteristic of the signal, the Mel-FrequencyCepstral Coefficients MFCCs of the wavelet channels are calculated. HiddenMarkov Models HMMs were used for the recognition stage as they give betterrecognition for the speaker-s features than Dynamic Time Warping DTW.Comparison of the proposed approach with the MFCCs conventional featureextraction method shows that the proposed method not only effectively reducesthe influence of noise, but also improves recognition. A recognition rate of99.3% was obtained using the proposed feature extraction technique compared to98.7% using the MFCCs. When the test patterns were corrupted by additive whiteGaussian noise with 20 dB S-N ratio, the recognition rate was 97.3% using theproposed method compared to 93.3% using the MFCCs.



Author: Mahmoud I. Abdalla, Hanaa S. Ali

Source: https://arxiv.org/







Related documents