Transcription Portlet

By | May 13, 2003

alphaWorks : Transcription Portlet

Transcription Portlet is 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.

Oh, I could certainly use one of those. Unfortunately, it requires WebSphere Voice Application Access. WVAA includes a set of plug-ins for WebSphere Studio (which is based on Eclipse) and a bunch of runtime stuff.

WebSphere Voice Application Access also provides the runtime components that make up the voice portal server infrastructure: WebSphere Portal Server (WPS), WebSphere Application Server (WAS), IBM SecureWay®, IBM DB2®, IBM HTTP Server, and others.

If you haven’t seen the price tag for WebSphere Portal Server, take my word for it, it ain’t cheap.

Now if someone would be willing to host this as a web service, that would be sweet. I could easily modify PhoneBlogger to have it send the recorded audio file to the transcription web service and then retrieve the text via HTTP, or even email, which the Transcription Portlet already supports. The submission of the audio file would obviously include a userid for a user who had already spent time training the speech recognition system on her voice.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.