In last week's article we took a look at the upcoming VoiceXMLPlanet Conference & Expo. In this part of our sneak preview we continue our exploration of the various sessions scheduled for the second day of the conference.
Day 2 (September 27)
As a technology standard, VoiceXML has done an excellent job of bringing thecomplex technologies around speech recognition, text-to-speech, telephony, callcontrol, etc. together and created new avenues for the development of interactivespeech applications. It doesn't stop there, however, as it's really the beginning of the ecosystem of applications, tools and services that can be created using voice.
As the founders and drivers of a company'svoice strategy, it is important to take a holistic view of what voice can dofor your business, the kind of products and services that you can provide to yourcustomers. In the keynote of Day 2, "Ozzie" Osborne will present the "BigPicture" and future of voice in the world of pervasive computing.
Whereas the first day is focused on technologies and fundamentals, day 2 isfocused around applications. Similar to the first day, the rest of the sessions of thesecond day are divided into two tracks - VUI/Best Practices track and the Real World VoiceXML Track.
VUI & Best Practices Track
It is typically said that eyeballs are important when you are developinginteractive e-business/web applications. Similarly, your ears are very importantwhen you are building interactive voice recognition applications. Yourapplication may be able to provide the most updated, real-time data from your backendlegacy systems using the flexibility offered by VoiceXML. However, if it doesn'tsound friendly and pleasant, you will probably not have a large number ofsatisfied users.& The "Writing for the Ear" session is tuned togive you tips and tricks about making speech applications fun and useful.
Typically as speech application developers we are presented with the challengeof taking the existing functionality which exists in today's web or non-webapplications, combining it with information about the user and converting into a usefulIVR system. As developers we are faced with critical questions such as "whatshould be speech enabled" and "what should be left?" "What conversational formshould be used - directed dialogs/mixed dialogs or natural language like?" "Whento use TTS?" The VUI Design Best Practices session will present the rules ofthumb in designing speech applications.
It is typically said that a speech application is only as good as the grammarit supports. However, developing efficient grammars can be a challenging task. Ontop of that, as developers we also need to make grammars modular in nature sothat they can be reused in different application scenarios. In the"Developing Better Grammars" session you can drill into some of the grammardevelopment standards, tools for grammar development and techniques foreffective grammar design.
When it comes to developing web applications, the layout, pictures and overall look and feel of the application is very critical in delivering a successfulweb-based application. In the world of Voice Applications, an easy flowingdialog interaction with the user is very critical. It is typically assumed thatwe need to build dialogs along the lines of Natural language like interaction e.g. "What can I do for you?" It has been illustrated that when it comes to dialog design, you don't always need to have a genie. It is quite possible to develop usable and easy to use interactions by carefully using directed dialogs. The dialog design evaluates the various approaches that a speech user interface design architect has and presents some useful guidelines.
Last (but definitely not the least) session of the day in the Best Practicestrack is about integrating VoiceXML and Voice over IP Networks. A combinationof VoIP and VoiceXML could bring the best of both worlds, the security and easeof setting up IP-based networks with the ease of use of speech-based interactions.