Europe Likely To Drive Multimodal Experience
With so many Europeans packing a mobile phone these days, network operators are finding it hard to deliver the revenue performances of past days. Value-added services like Wireless Application Protocol (WAP) were supposed to deliver the needed top line growth, but WAP has had difficulty achieving mass market acceptance thus far. So mobile operators are looking at voice portals and other voice-driven services as a solution to these problems. But don’t think the mantra to consumers will be “forget WAP, try speech instead”. Wireless analyst, Rupert Reid of the ARC Group, believes that mobile operators will position voice alongside WAP. “Speech recognition is a useful interface more than a killer app…multimodal is where European operators are going.” Daniel Hawkins of the Voice Business Group at Datamonitor sees development on the horizon a little differently. “I would like to predict full multimodal, but I’m not convinced. I think we will see more apps where you phone in and SMS back, or vice versa.” Voice Web Consulting (VWC) agrees with both perspectives and summarizes the operator charter as multimodal being the destination and multi-access, the port of departure. There are several reasons Europe is likely to lead the rest of the world in this journey. To begin with, network operators want to protect their investment in WAP. While mobile providers over promised and under delivered, the facts are that consumers use text-based services (WAP and SMS). Many of the same market drivers for WAP apply to voice portal services as well. Furthermore, the European telecommunications industry has a tradition of evolving around standards. All European wireless networks operation using the GSM (Global Systems Mobile) air interface and this trend of standardization will continue with 3G. The European Telecommunications Standard Institute (ETSI) drives much of the stardards development effort. In fact, they have a working group developing a standard for distributed speech recognition in mobile networks where the front-end feature extraction process occurs on the phone and is then compressed and sent over a data channel (vs. today’s voice channel) to the server for back-end speech recognition. The project is called Aurora. Europe is also ahead of others with regard to plans for 3G wireless networks. Most European countries have already auctioned off spectrum licenses and the UK is in a friendly race with Japan to bring higher bandwidth trial services to market. While multi-access/multimodal is technically not dependent on 3G, the bandwidth needed to support video will only come with 3G. Reid believes that speech will drive the demand for 3G data services. Two issues stand in the way of multimodal development. One is the sheer complexity of user interface and application design. Today an application developer can create functionality to support a user accessing and using both a voice user interface and WAP access interface. But with multimodal, the interaction is no longer dictated by the application, but rather by the network. The network responds in real-time to the desires of the telephony user while surfing. The second inhibitor is one that mobile operators are intimately familiar with…handsets. Today’s battery technology prevents simultaneous use of voice and data channels. This isn’t a problem for telematics in the car though. For all these reasons, multimodal will evolve gradually…from single multi-access applications to multimodal sessions in open, standards-based networks. Vectoring this progression will be the increased development of VoiceXML-based voice sites, leveraging the back-end logic of current Web sites.
Kathy Frostad is principal advisor for Voice Web Consulting (VWC), a specialty-consulting firm solely focused on voice recognition technologies and their business application in wireless, enterprise, Internet, and PSTN environments. For more information, visit them at http://www.voicewebconsulting.com.