The division into segments is done using a specially modified speech recognizer.
It has two essential components that do not exist in a text dialog system: a speech recognizer and a text-to-speech module.
The major steps of a typical speech recognizer are as follows:
Each incoming response is then processed automatically by the speech recognizer that has been optimized for non-native speech.
It will then start up a speech recognizer, giving it a speech recognition grammar.
The speech recognizer takes about 12 times longer than the duration of a given message to produce a transcript.
He was well known for his oft-quoted quip that "Every time I fire a linguist, the performance of the speech recognizer goes up".
His work in speech synthesis led to ideas of how to create a single-chip speech recognizer.
The new owner must also introduce the speech recognizer to on-the-job jargon.
The people who telephone E-Trade and decide to buy stock through a speech recognizer instead of a person deal with what is called a speaker-independent program.