Speech Recognition

Course: 
AttachmentSize
speechrecognition.pdf1.35 MB
immediate translation program using speech recognition techniques136.26 KB
Description: 

 

Al-Najah national university

Information Technology

Computer Information System

 

Speech Recognition

 

Supervision: Dr.Baker Abd Al-haq

 

Submitted by:

 Majd Fawzi Fatouh

 

   2010
Abstract

                                    

Accurate recognition of spontaneous speech is one of the most difficult problems in speech recognition today, so I prepare these papers.

To convert speech to on-screen text or a computer command, a computer has to go through several complex steps. When you speak, you create vibrations in the air. The analog-to-digital converter (ADC) translates this analog wave into digital data that the computer can understand. To do this, it samples, or digitizes, the sound by taking precise measurements of the wave at frequent intervals. The system filters the digitized sound to remove unwanted noise, and sometimes to separate it into different bands of frequency (frequency is the wavelength of the sound waves, heard by humans as differences in pitch),It also normalizes the sound, or adjusts it to a constant volume level. It may also have to be temporally aligned.

Next the signal is divided into small segmentswhich are called frames, then these frames pass throw a technique called windowing is used to achieve this result. Windowing allows the portion of the sample that is closest to the center of a window to be more heavily weighted than the parts of the sample that are further away from the center of the window.

The next step is using powerful and complicated statistical modelingsystems. These systems use probability and mathematical functions to determine the most likely outcome. The most common model is hidden marcove model,which describe a two-stage stochastic process.

Finally using traning method in order to get the best path of the training data .