|
Speech recognition technology is currently prevalent in customer service applications such as directory assistance, the provision of flight information, stock trading and catalog order taking. A major barrier to the widespread adoption of speech recognition technology in the IT setting is the current lack of open speech standards. The leading contenders are Voice Extensible Markup Language (VoiceXML), now available in a version 2.0 format, and SALT, a newer standard sponsored by the SALT Forum and designed to support multimodal applications. Technically, both of these schemes are presentation layer languages that provide a familiar Web development platform, support flexible deployment architectures, and afford the possibility of application portability. The technology remains complex, however, with the time required to deliver an average speech application ranging from 3,000 to 6,000 person hours. Nevertheless, standardization will have to make the development and implementation of speech recognition systems easier, more flexible and less time consuming before it becomes an enterprise-ready technology.
|
Information Sources, Inc.
![[Get Copyright Permissions]](http://license.icopyright.net/images/icopy-w.gif){kind=small}