Automatic Speech Recognition System

Figure 1.3 shows the Automatic Speech recognition systems be able to be classified as follows Figure 1.3 Automatic Speech Recognition system classification.

1.5 Merits and demerits of speech recognition
Merits
• Speech is a very normal way to communicate, and it is not need to sit at a keyboard or work with a remote control systems.
• No training and testing phases required for users.
• Avoid handwriting and spelling troubles.
Demerits
• Still the best speech recognition applications sometimes make errors. They raise noise or some other sound the number of errors will increase.
• Speech recognition works best if the microphone is close to the user.More distant microphones like on table, wall will tend to increase the number of errors.
• Speaker

1.7.1 Pronunciation differences
In Speech application system there are many issues to be raised. Such as when the speaker communicate with the speaker means they pronounce a single word can be reached to the speaker in many ways. So another speaker recognizes the word become confused and misineterpt. The same situation can be occurring in the slower and quicker communications
1.7.2 Identifying the context of the words
Now a days many software’s will be developed. Like the computer software’s not capable to recognize the anticipated meaning of a group of words and also number of problem will be raised in the transcribed text. For example “week” and “weak” the two words have a similar sound but the meanings are different. Those reasons raised in the accurate punctuation are almost not possible for the software to place based exclusively on knowing the continuous words [4] [5].
1.7.3 Speaker