Speech Recognition

Read Complete Research Material



Speech Recognition

Speech Recognition

Speech Recognition

Speech recognition is also termed as voice recognition. It is an alternative typing to the keyboard. It is useful for individuals who are physically disabled. It is the process by which computer understands the spoken words and translate into written words. It is the way of talking to the computer. This technology is getting modern day by day. Technology is advancing every day. Each day brings new ideas, revolutions and inventions. Today, no house and organization, is without computer. Information Technology is playing a vital role in reshaping lives. The use of Internet in business organizations, offices, houses and other work places bring comforts to the lives of people. In business organizations, need of presentations and research work cannot be completed without Internet. The use of Internet for any purpose also brings risks and challenges to the organizations. Computers are too obvious today that they understand our speaking through software which is mainly speech recognition software. Computer understand that individual speak in the same voice and tempo. There are speaker independent systems which enables the computer to understand the talking of various speakers. There are SR systems, which enables a computer to recognize our words. Similar vocabularies and words are easier to understand by computer. The vocabulary may have thousands of meanings of words. The entire ASR systems have the accuracy of about 98%. Speech recognition enables a user to input computer through voice.

Speech Recognition Software

Speech Recognition software is used by the general public. Individuals who are slow in typing, and finding difficulty in typing show their interest in speech recognition software. The user tries to contact with the system and replies to the commands and instructions. The use of this software requires training before its use. Training is often called as enrollment. The major part of the software is the translation part. There is much speech recognition software, which break down the spoken words into phonemes, and the basic sound from which the words and syllables are built up. In order to work speech recognition software perform efficiently, there is a need to train the system to understand the user's pitch and quality of sound.

Steps involved in Speech Recognition Software

Enrollment

The voice quality and pitch of individuals differs from the others. The first step involves in speech recognition is to read the article displaying on the screen. This process is referred to as training or enrollment which creates numerous files and tells the software that how to speak. This process takes less than ten minutes. New speech recognition software does not require this enrollment. It is also essential in order to get the best results. This training has to be done at once and after that software is set up to perform as needed.

Audio Input

The first and the foremost thing of using this software involve the audio which come into the recognition engine. This speech does not only contain the data but also the noise of the background. The job of the speech recognition is to convert the ...
Related Ads