Automatic Speaker Recognition using Mel Frequency Cepstral Coefficients (MFCC) and Fast Fourier Transform (FFT)
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Msila
Abstract
This work focuses on automatic speaker verification identity via his voice, which
is a recent task of the Automatic Speaker Recognition field. This work is divided into
two phases: The first phase "called training or learning phase" consists of acquiring
and saving audio recordings of a group of speakers into PC. In the second phase
"testing phase", an anonymous speech is introduced and compared its characteristics
with the characteristics of previous recordings, using a set of algorithms (such as:
MFCC, FFT) to make a decision. The speech files that were recorded and processed in
this work are obtained using a personal computer microphone instead of special
equipment, which explains the weakness of the obtained results.