Extending the vocabulary of a client-server speech recognition system

摘要

A network spoken language vocabulary system for a speech application comprises: a network server (12) and a plurality of network clients (14, 16, 18) communicating with one another over a network; a central vocabulary list (24) in the server for recognizable words; and a speech recognition engine (32) and a local vocabulary list cache (36) in each of the clients. The cache (36) has therein previously recognized words communicated from the central vocabulary list (24) in the server and new words recognized by the speech application. Each of the new words is communicated to the server and added to the central vocabulary list and each of the new words added to the central vocabulary list is communicated to at least one other of the clients for addition to the cache. The new words can be automatically communicated to and from the server. <IMAGE>