Professor Wahlster provides an invaluable overview of research in European universities and company laboratories in the areas of speech, multimodal, and multilingual technologies, looking forward to technologies that will be available in the market place within two years. His keynote will help attendees develop plans for the types of applications they will be able to build and deploy in the future.

Professor Wahlster is the Director and CEO of the German Research Center for Artificial Intelligence and a Professor of Computer Science at Saarland University (Saarbrücken, Germany). In addition, he is the Head of the Intelligent User Interfaces Lab at DFKI. He has published more than 170 technical papers and 8 books on language technology and intelligent user interfaces.

Experienced developers present their key learnings and insights for developing speech applications for the real world. Brian Budd describes an approach to design documentation that has successfully been used during the implementation for a major UK financial institution. Michael McTear explains the design of an application for diabetics to use the telephone to report their readings for weight, blood sugar and blood pressure on a regular basis with speech or touchtone input and to receive feedback on users’ conditions.

Statistical Language Models (SLM) are an alternative to the context-free grammars frequently used by speech recognition systems. What are the pros and cons and the strengths and weaknesses of SLMs? Are SLMs really natural language processing? How expensive are SLMs to create and maintain? How to implement SLMs and make them work efficiently? How can the dialogue manager recover when an SLM fails to recognise the user’s utterance correctly?

With the improvement in quality of human-like voices generated by text-to-speech synthesis engines, new applications and uses are emerging. This session reveals how you can put speech synthesis to work in new ways. Learn how one company uses a synthesised voice as its brand. Discover how Interactive Digital Assistants created by integrating synthesised voice with an avatar can be accessed both via desktop computers and mobile devices. Find out how Braille can be generated from text, making voice dialogues available to visually impaired users.

Learn how synthetic voices can be personalized to sound like a specific person, and what new applications can be enabled by personalized synthetic voices.

Telephone recordings, email, chat, surveys, and tweets, all contain valuable information about how your organisation performs. This session will reveal how to capture and analyse information from multiple formats and media to improve business. Learn the pros and cons of various analytics technologies and how the right combination of technologies delivers higher accuracy, more efficiency and better intelligence. Discover how to analyse this information and determine what actions to take to improve business practices.

SpeechTEK Europe’s Avatar Challenge presents a collection of newly-developed avatars designed for a range of business applications. Find out the winners of the ‘expert's prize’ and ‘people's choice’ in SpeechTEK’s Avatar Challenge.

Professor Wahlster provides an invaluable overview of research in European universities and company laboratories in the areas of speech, multimodal, and multilingual technologies, looking forward to technologies that will be available in the market place within two years. His keynote will help attendees develop plans for the types of applications they will be able to build and deploy in the future.

Professor Wahlster is the Director and CEO of the German Research Center for Artificial Intelligence and a Professor of Computer Science at Saarland University (Saarbrücken, Germany). In addition, he is the Head of the Intelligent User Interfaces Lab at DFKI. He has published more than 170 technical papers and 8 books on language technology and intelligent user interfaces.

Testing and tuning of speech applications are vital for good customer experiences. Darren Chapple will explain how to integrate testing into the overall project plan, how to minimise the impact of outside influences on testing, and how to improve testing efficiency. Stuart Walsh addresses the new problem of testing and tuning systems using VoIP. Learn how to measure end-to-end voice quality, how to test VoIP networks and assure their scalability, and how to improve the overall technical performance.

The education sector presents one of the biggest opportunities for speech technology. This session reveals how the use of speech technology services delivered on mobile phones can provide educational services to millions who would otherwise be underserved. A fascinating case study demonstrates how speech recognition technology can be combined with other criteria to assess the speaker’s pronunciation quality. Learn about cloud computing architecture and the challenges with mobile solutions, business problems, use cases, business models, marketing challenges, and the problems of integrating speech with visual interfaces.

These presentations describe three new technologies that have the potential to improve business practices and make business more competitive. Learn how high-definition voice provides businesses with a crystal clear, high-quality real-time conversation. Discover a new approach to spontaneous speech transcription and how it can improve real-time conversation. Review various methods, philosophies, and related pros and cons about how to implement speaking avatars on the Web, kiosks, mobile applications, embedded devices, and live TV shows.

SpeechTEK Europe’s Avatar Challenge presents a collection of newly-developed avatars designed for a range of business applications. Find out the winners of the ‘expert's prize’ and ‘people's choice’ in SpeechTEK’s Avatar Challenge.