IBM Launches Transcription Portlet

May 02, 2003

IBM has launched Transcription Portlet, a voice portlet for transcribing telephony-based dictation. This technology makes large-vocabulary speech recognition technology available to telephony-based portal applications (portlets). It provides Java APIs through which developers can integrate transcription capabilities into a portlet application. Transcription portlets enable a user to call into a portal, dictate information, and then have the dictated audio transcribed into text. The results of the transcription (the transcribed text) can then be accessed through the Internet or sent to the user by email. The transcription portlet, available as a free, trial download on alphaWorks enables users to: call into a portal, dictate information, and then have the information converted into text; create new user accounts; register an email address where text can be sent; view enrollment information; and correct transcribed text. In addition, the transcription portlet is speaker-dependent. While this does mean that individual training is required, it also increases the accuracy for general dictation and allows for a larger vocabulary. IBM is continuing to explore new methods of extending the voice capabilities interacting with its WebSphere Portal. IBM recently introduced WebSphere Voice Application Access, which is designed to make it easier to build voice portals and allow internal and external customers access to the same applications through a voice interface.