Hey Speech-Heads:

This SpeechTEK Europe Press Release is straight from the Horse’s Mouth:

FOR IMMEDIATE RELEASE:

Leading speech synthesis developer, CereProc–recently the subject of global media attention, including coverage on the Oprah Winfrey Show, for their development of a synthetic voice for film critic Roger Ebert–will reveal the latest innovations in speech synthesis at London’s SpeechTEK Europe Conference.

Paul Welham, Cereproc’s Chairman & Chief Executive Officer comments, “There is something special about speech. A person’s voice is not just a means of communicating; it also gives a deep impression of who we are. Humans use their voices to show emotion, persuade, and convince, to calm and to excite. So, what can we do if a person loses the power of speech? The answer lies now with advanced speech synthesis technology.

“Roger Ebert, arguably America’s most famous film critic, lost the ability to speak after a thyroid cancer operation. Although he used speech synthesis available on his PC to communicate, he was deeply frustrated because the voice did not sound like him. Using Roger’s commentaries from DVDs, CereProc created a voice that mimics his original speaking style, including emotional qualities. Roger can now manipulate his own voice again, to alter pronunciation, pitch, intonation, and reflect his personal voice qualities.”

Cereproc are just one of the participants in a Panel Session on New Uses of Text-to-Speech Synthesis on the second day of SpeechTEK Europe.

Other Conference highlights include:

• Transport for London on using automated voice technology to help customers make over 24 million trips a day, and how to minimise the impact of technology failures.

• Reducing fraud and replacing out-dated methods of authentication with the latest in voice biometric systems.

• Will it work in Swahili? Understanding how to design, implement and test multilingual speech applications to work across multiple languages and cultures.

• Real world applications: a major UK bank on successful design documentation; and how the ‘phone is being used to report diabetic readings and feedback patients’ conditions.

• Cutting edge advances in high definition voice, spontaneous speech, avatars – and how they are improving business performance and competitiveness.

This first-ever SpeechTEK Europe offers a global perspective on speech technology and its future, discussing and showcasing innovations, applications, services and solutions for those considering or already using this powerful range of technologies to improve customer service, increase business performance, and boost the bottom line.

But, whilst development and adoption of speech technology has been rapid, barriers and challenges still remain for the speech industry, and these are also addressed at SpeechTEK Europe.

Caroline Leathem, Interaction Specialist at Verizon comments, “Overcoming the negativity associated with speech automation in contact centre solutions is a major challenge. There are still too many poor designs out there and if we, who understand the limitation of automated voice solutions, are frustrated by an IVR then you can bet your bottom dollar most other people will be. By recognising the importance of interaction design skills, planning regular usability test cycles, and making the technology do the work rather than the caller, we can increase people’s confidence in speech technology as a whole.”

Loquendo’s Rosanna Duce believes more work is needed on usability: “The barriers are that choosing speech technology to solve a business issue – be it entrusting users to a voice-enabled IVR, adopting TTS as the voice of your device, or choosing a voice-dialling application–is still a difficult task and the reason for this is that the ‘usability’ of these applications is not often as ‘user friendly’ as they ought to be.”

Conference Chair, Jim Larson, believes that standards remain an area of concern for the industry–and this will also be on the agenda at SpeechTEK Europe: “Development language standards [are needed] so that applications developed for a mobile device from one manufacturer can be used on devices from other manufacturers. VoiceXML 2.0 achieved this for voice-only applications. VoiceXML 3.0 and related W3C standards will hopefully do this for multimodal applications.”

The full SpeechTEK Europe programme is available HERE, along with registration information, details of registration savings, and free entry exhibition tickets.