Voiceprint Feature Extraction Based on Wavelet Packet Decomposition

The main purpose of speech signal recognition system research is to improve the recognition rate of speech signal and reduce the recognition time. The feature extraction of speech signal in the process of recognition is one of the key aspects. In this paper, the speech feature extraction is to use the auditory characteristics of the human ear to decompose the wavelet packet into five levels and extract the dynamic features contained in the frame signal. After further processing, the speech feature parameters (DWPT parameters) are obtained. Simulation shows that the speaker recognition rate has been significantly improved compared with other traditional methods.

[1]  S. X. Chen,et al.  Initial codebook method of vector quantisation in Hadamard domain , 2010 .

[2]  Dipti B. Shah,et al.  HMM-Based Lightweight Speech Recognition System for Gujarati Language , 2018 .

[3]  H. Gish,et al.  Text-independent speaker identification , 1994, IEEE Signal Processing Magazine.

[4]  Ronald W. Schafer,et al.  Theory and Applications of Digital Speech Processing , 2010 .

[5]  Daniela Schiopu,et al.  Using Statistical Methods in a Speech Recognition System for Romanian Language , 2013, PDeS.

[6]  Rosemary T. Salaja,et al.  A Life-Based Classifier for Automatic Speech Recognition , 2014 .

[7]  Max A. Little,et al.  Novel Speech Signal Processing Algorithms for High-Accuracy Classification of Parkinson's Disease , 2012, IEEE Transactions on Biomedical Engineering.

[8]  James R. Glass,et al.  Speech feature denoising and dereverberation via deep autoencoders for noisy reverberant speech recognition , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[9]  Kandarpa Kumar Sarma,et al.  Machine learning based sample extraction for automatic speech recognition using dialectal Assamese speech , 2016, Neural Networks.

[10]  Ravenswood Ave,et al.  Phone-based Cepstral Polynomial SVM System for Speaker Recognition , 2008 .

[11]  Chin-Chen Chang,et al.  A fast LBG codebook training algorithm for vector quantization , 1998 .

[12]  Haizhou Li,et al.  An overview of text-independent speaker recognition: From features to supervectors , 2010, Speech Commun..

[13]  R. P. Ramachandran,et al.  Robust speaker recognition: a feature-based approach , 1996, IEEE Signal Processing Magazine.