Speaker recognition

Abstract

Firstly, I would like to express acknowledgement to Prof. Nikos Fakotakis, who served as a supervisor of my Ph.D. study. His comprehensive support I enjoyed from the very first day of my work at the Wire Communications Laboratory. It was his countenance, which made possible the successful completion of my study.I would like to express gratitude to Prof. George Kokkinakis, whose profound analysis of my prospective submissions helped me to improve both the presentation style and overall quality of the manuscripts. Subsequently, I would like to thank to: Assoc.Prof.John Mourjopoulos who directed my attention to the psychoacoustic aspects of speech perception, Ass.Prof. Evangelos Dermatas who inspired my interest to the recurrent neural networks, and to Prof. Michael Vrahatis who initiated me in the evolutionary computation techniques. Further, I would like to express thanks to: Dr. Anastasious Tsopanoglou, with whom I had regular discussions during the first year of my study and to Dr. Ilyas Potamitis, whose insights helped me to avoid many snares during all these years. In addition, I would like to thank to all colleagues who contributed to the comfortable collaborative atmosphere at the Wire Communications Laboratory. Finally, I would like to express acknowledgement to the State Scholarship Foundation of Greece (IKY), which financially supported my Ph.D. study during the years 2002/2003/2004/2005. The IKY scholarship gave me the placidity to focus on my study, which I enjoyed very much. I highly appreciate this support.v Contents: Dedication ……………………………………………………………………….. iv Acknowledgements ………………………………………………………………. v Contents ………………………………………………………………………….. vi List of figures …………………………………………………………………….. x List of tables ……………………………………………………………………… xiv List of abbreviations ……………………………………………………………… xv Notations and operations ………………………………………………………… xviii PART I. INTRODUCTION TO THE SPEAKER RECOGNITION TECHNOLOGY AND OVERVIEW OF THE STATE-OF-ART

Kajian Penelitian Pemrosesan Bunyi dan Aplikasinya pada Teknologi Informasi

“…Metode MFCC menjadi populer karena mampu meniru kemampuan manusia dalam menangkap ciri suara atau bunyi. Penggunaan MFCC dapat dilihat pada beberapa penelitian, seperti pada [18], [19], [20], [21].…”

“…Pada Gambar 1 menunjukkan tahapan yang dilakukan pada MFCC [18]. Setiap tahapan memiliki tujuan dan pemrosesan yang berbeda.…”

Mel-Frequency Cepstrum Coeffficients as Higher Order Statistics Representation to Characterize Speech Signal for Speaker Identification System in Noisy Environment Using Hidden Markov Model

“…Compute the J cepstrum coefficients using Discrete Cosine Transform Compare to other feature extraction methods, Davis and Mermelstein have shown that MFCC as a feature extraction technique gave the highest recognition rate (Ganchev, 2005). After its introduction, numerous variations and improvements of the original idea are developed; mainly in the filter characteristics, i.e, its numbers, shape and bandwidth of filters and the way the filters are spaced (Ganchev, 2005).…”

“…After its introduction, numerous variations and improvements of the original idea are developed; mainly in the filter characteristics, i.e, its numbers, shape and bandwidth of filters and the way the filters are spaced (Ganchev, 2005). This method calculates the cepstral coefficients of a speech signal by considering the perception of the human auditory system to sound frequency.…”

“…Block diagram of the method is depicted in Figure 4. For more detailed explanation can be read in (Ganchev, 2005) and (Nilsson, M & Ejnarsson, 2002). After a process of windowing and Fourier transformation, performed wrapping of signals in the frequency domain using a number of filters.…”