A text-dependent speaker-recognition system

Abstract

In this work, a voice recognition approach is developed and presented. It is a based on acquiring the signal of vocal cords' vibrations of a person using a piezoelectric transducer element attached on a collar wrapped around the neck. The recognition is then based on the vocal cords vibrations' pressure of the individuals and not their normal voices. Due to the varying nature of the collected signal, the analysis was performed by applying the Short Term Fourier Transform technique to decompose the signal into its frequency components. These frequencies represent the vocal folds vibrations' frequencies (100-1000 Hz). The features in terms of frequencies' interval are extracted from the resulting spectrogram. Then, 1-D vector is formed for identification purposes. The person's identification is performed using the correlation coefficient as a similarity measure. The results show that a high percentage of recognition is achieved and the performance is much better than many existing techniques in the literature.