How speech recognition works
An application can have different types of user interface (UI), and controlling the UI using speech is one of the approaches of user interaction. Using the speech recognition system, users say what they want and the computer executes the command and the results are reflected on the UI.
We can categorize the patterns of speech recognition in the following two ways:
Command mode: This is the mode where you say a command and the speech recognition engine recognizes the speech. As an example, you may want to start and stop a game by just saying "start" and "stop".
Sentence mode or diction mode: This is the mode where you can say a sentence to perform an operation. As an example, to rotate a line you can say "rotate the line".
At the first glance, speech recognition looks like a simple matching logic, but indeed it is not. The speech recognition engine consists of the following two major modules:
Acoustic model
Language model
Each one of the modules has a sole responsibility...
