septendesilyon
Newbie level 4
Hi everyone,
I have a project that related to speech recognition. Firstly, I searched some sources than decided to use MFCC and Artificial Neural Networks. In this algorithm we must train the system with some voices but our project leader does not want it. For example, in Nokia Phones we create "Micheal" in telephone directory, when we speak out to "Micheal" the phone automatically call the Micheal. In this example we did not train the phone before saying Micheal. Anybody know how does this system work ? I assume that the phone creates text to speech than compare this speech with user voice. We will create Voice Recognation System like Nokia.
Thanx a lot.
I have a project that related to speech recognition. Firstly, I searched some sources than decided to use MFCC and Artificial Neural Networks. In this algorithm we must train the system with some voices but our project leader does not want it. For example, in Nokia Phones we create "Micheal" in telephone directory, when we speak out to "Micheal" the phone automatically call the Micheal. In this example we did not train the phone before saying Micheal. Anybody know how does this system work ? I assume that the phone creates text to speech than compare this speech with user voice. We will create Voice Recognation System like Nokia.
Thanx a lot.