Title :
Procedure for Cepstral Analysis in tracing unique voice segments
Author :
Banerjee, P.S. ; Chakraborty, Baisakhi ; Banerjee, Jaya
Author_Institution :
Dept. of Comput. Sci. & Eng., Jaypee Univ. of Eng. & Technol., Guna, India
Abstract :
The main motive in venturing into Cepstral Analysis of speech is to understand the minute subtle details of Cepstral Analysis mechanism. Another unique aspect that we were interested in performing vocal tract and source information separation by Cepstral Analysis and hence understanding liftering concept in cepstral Analysis. In basic spectrum analysis the range of frequency specific to a particular voice sample is obtained where as we decided to device a unique pitch bifurcation method by Cepstral analysis. In the process we were also considering the formant aspect of the speech and to get that information we took the help of Cepstral analysis. At the end we will also try to differentiate real cepstrum from that of complex cepstrum.
Keywords :
cepstral analysis; speech processing; cepstral analysis mechanism; source information separation; spectrum analysis; unique pitch bifurcation method; unique voice segments; vocal tract; Cepstrum; Feature extraction; Frequency-domain analysis; Speech; Speech processing; Speech recognition; Cepstral Analysi; cepstrum; liftering;
Conference_Titel :
Computing for Sustainable Global Development (INDIACom), 2015 2nd International Conference on
Conference_Location :
New Delhi
Print_ISBN :
978-9-3805-4415-1