Abstract:
Since the old time, sound is considered an important mean of communication between human beings. Human-Computer Intelligent Interaction (HCII) is an emerging field of science it is argued that for the computer to be able to interact with humans; it needs to have the communication skills of humans. One of these skills is the ability to understand the emotional state of the person.
One expressive way humans display emotions is through vocal expressions, in this work a new concept is provided which is using the voice to determine the emotion of the person. The idea of the project is to create a simulation program that will apply in a mobile telephone to determine the emotion of the person.
Emotion recognition is a new field and one of a funny and interesting branch in the artificial intelligence. By using this type of recognition system machine (mobile telephone) can recognize the emotion of the person during the call; two emotions were selected happiness and anger.
This program regard as step in this field to establish integrated systems to recognize emotions. The steps which used to recognize the emotion of the person was done through three phases:
Phase one was performed by recording speech of four persons one is male and the other are female using MP3 at sampling frequency 41100 Hz and 8 bits resolution ,and then select some words individually from this speech by using audio program. The speech was recorded in two modes happiness and anger.
In phase two, Spectral analyses were done by using Fast Fourier Transform (FFT).
In phase three, the output of phase two compared with reference voices stored as a database.
Matlab is used in this research for containing it many tools for signal processing and tools for comparing, beside that Matlab is a simple program.
