VoiceXML Review - Letter from the Editor

Volume 2, Issue 5 - July/August 2002

July / August 2002

Dear Readers,

The theme of our July/August VoiceXML Review is on interfacing to speech resources. We are pleased to present you with an assortment of excellent in-depth technical articles on this subject, written by technologists who play leading roles in the specification and development of speech resource interface technologies.

In the past year, there has been a fair share of excitement centered on the MRCP (Media Resource Control Protocol) Internet draft submitted to the IETF by Cisco, Nuance and Speechworks. The idea is to provide a standard interface to speech resources (speech recognition, text-to-speech synthesis, and speaker verification). Those of us who have experienced the pain of integrating speech resources from multiple speech resource vendors (each having its own proprietary interface) into our voice platforms can't help but get enthusiastic about such a concept!

Recently the IESG has chartered the Speech Services Control Working Group (speechsc) of the IETF to formally to specify this interface. In his article entitled "The IETF Speech Services Control Working Group.", Eric Burger, co-chair of the speechsc working group, provide us an update of the working group's vision and progress to-date.

On the text-to-speech synthesis front, quite a bit has changed in the W3C Voice Browser Working Group's Speech Synthesis Markup Language (SSML) draft specification since we last updated you on this subject. In his article, "Update on SSML", Dan Burnett, co-editor of the specification, gives us a technical walk-through of what has changed, and what to expect moving forward.

Complimenting Dan's SSML article, in our regular Speak-n-Listen column this month, Matt Oshry walks us through how to properly render the result from a built-in VoiceXML date field using a combination of SSML and ECMAScript, as well as another approach using pre-recorded audio. The technique presented can be readily applied to rendering speech/audio from other textual formats as well. To have your VoiceXML question answered in a future column, drop Matt a note at .

Finally, Rob Marchand is back with an "eventful" First Words column. Events in VoiceXML are perhaps the least understood part of the language, particularly among beginners. However, events are an integral part of the language, and in order to write good VoiceXML applications you are going to need them. While Rob's columns in the past have touch on events, this column goes into a bit more depth and provides an excellent opportunity to round out your understanding of this area.

Sincerely,

Jonathan Engelsma
Editor-in-Chief, VoiceXML Review

back to the top