29-01-2009, 11:43 PM
Serving as a bridge between people and computer databases, interactive voice response systems(IVR) connect telephone user with the information he need, from anywhere at any time. IVR turns a telephone into easy-to-use computer terminal, enable caller to access specific information from a host computer database without going through a live operator. The quacking pace of adoption of speech solutions in IVR industry is currently driven by improvements in speech algorithms, natural language processing, vocabulary management and language modelling. Automatic Speech Recognition solutions enables customers and prospects to conduct a myriad of transactions by interacting with the computers in the simplest possible way by speaking over the telephone. The goal of speech-enabled applications has always been to assist the callers in obtaining information and performing transactions simply by speaking naturally. Speech recognition can deliver more than a simple upgrade or enhancement of touch-tone IVR applications. Automatic Speech Recognition (ASR) is a technology that allows a computer to identify the words that a person speaks into a microphone or telephone. This seminar deals with Large vocabulary continuous speech recognition. The main components of the system are front-end parameterization, acoustic modelling, language modelling and decoding. Current state of the system is also mentioned. Finally it deals with the features and applications of the system.