Archive for October, 2007

Impressions from Voice Days 2007 in Bonn, Germany

Sunday, October 21st, 2007

Last Wednesday and Thursday I went to Voice Days 2007. This yearly event is held in the historic setting of the former German Bundestag, on the banks of the river Rhine in Bonn, only 250 km from Brussels. Despite its English-sounding name, the first purpose of the conference is to bring together speech application professionals from the German-speaking countries, i.e. Germany, Austria and Switzerland in the first place. So a decent knowledge of German comes in handy if you want to follow the debates.

Voice Days 2007 in Bonn, Germany

According to the organizers, this year’s event assembled more than 50 exhibitors, as well as more than 900 registered participants. This makes Voice Days the biggest speech technology event in Europe, larger than its French-speaking counterpart Vocal Expo.

Given the cultural aspect of customer-facing speech applications and the obvious need for local knowledge and partnerships, this fragmented market approach is not very likely to change in the near future. Of course, responsiveness to local needs does not necessarily equal parochialism and monolinguality. For example, speech-driven public transport information lines in München and Hamburg do support one or more foreign languages. And in Switzerland, multilingualism is the rule rather than the exception. The MeteoSwiss weather information line, for example, is available in four languages. So multilingual support is not to be seen as an exceptional feature, entailing extra costs: it’s more and more becoming a basic requirement, even in countries that are otherwise officially monolingual.

Voice Days 2007 in former German Bundestag, Bonn, Germany

As one might expect, the German speech technology industry is extremely well organised. Initiative Voice Business, the organization behind Voice Days, also drives the German-speaking Voice Community. This competence network for speech-enabled dialog systems plays a pivotal role in market development through market reviews, technology sharing, white papers, tutorials, demo portals, and, last but not least, a near-free Voice Testcenter aimed at application developers. One excellent example of this industry-wide collaboration is a thorough 100 page guide (in German) on “Quality criteria, measures and procedures for speech applications“, which was distributed to all conference participants. Other related initiatives are the yearly Voice Award for the best applications, and Voice Contest, which awards technical or business innovation in the local speech dialog industry. Since 2004, more than 150 German-speaking speech-enabled services have been tested under these programs.

Another notable – yet private – market review initiative is Voice Compass. The current version focuses on the German-speaking market, but the 2008 edition of this compendium aims at an international public and will be published in English.

The river Rhine in Bonn, Germany

On the application side, I learned about the existence of no less than three public transport timetable services, available respectively in Munich, Hamburg, and Nordrhein-Westfalen. At the Telenet booth, I saw innovative speech-enabled crossword puzzles and took football penalties on a web screen by shouting “boom” into a phone. I heard live demos of callers getting automatically classified by age and gender (Siemens), or by mother tongue (T-Mobile). Crealog demoed an Interactive Voice Video Response (IVVR) application where callers select television programs on their mobile phone using voice commands. Quite a technical feat, but I wasn’t completely convinced by the ergonomy of the user interface.

Keynote speaker Geraldine Wilson from Yahoo! Europe predicted that search will be the doorway to the mobile Internet just as it has been on the good old Internet 10 years ago. Unfortunately she did not elaborate on the role that speech input would play in this trend. Ms. Wilson pleaded strongly for  physical search button on each mobile phone – which made me think of Nuance’s plea for a physical speak button at their last European Conversations conference. Interesting to see that physical – read: hardcoded – features are still considered that important in a technological world driven by personalization, virtuality and constant change. Why not get rid of all physical buttons altogether, and turn the mobile phone into a multi-modal portal-like personalisable device, with freely downloadable configurations and skins? Ms. Wilson repeatedly mentioned Apple’s iPhone, but remained silent on Google’s gPhone. Which, admittedly, is currently also known as the vaporPhone.

Frederik Durant on the banks of the Rhine in Bonn, GermanyÂ

One way to measure the maturity of a local speech dialog market is to have a look at the investments made in managed services. Companies like T-Com, VoltDelta, Crealog or D+S Solutions have been offering managed voice portals for several years now. In the mean time, speaker verification has been added to various product offerings.

Given Germany’s undisputable leading role in the European speech dialog business, it is surprising to see how few Belgian telecom technology managers or decision makers made the effort of travelling to nearby Bonn. The lack of a high-speed train connection should be no excuse for missing out on a valuable conference like Voice Days. Let’s hope this post has helped a bit to bridge the culture gap. See you in Paris then, for Vocal Expo 2008?

Job ad: Technical engineer (media services) with VXML experience at Pauwels Consulting, Zaventem, Belgium

Monday, October 8th, 2007

See the original job posting at Stepstone.