1 / 15

Paul Welham (paul@cereproc) Cere Proc Ltd 10 th March 2008

Paul Welham (paul@cereproc.com) Cere Proc Ltd 10 th March 2008. Agenda. Purpose of today’s presentation Cere Proc - Background Cere Proc - What makes us unique Cere Proc’s CereVoice and the need for character and emotion in synthetic voices Summary. Purpose Of Today’s Presentation.

briar-cummings
Download Presentation

Paul Welham (paul@cereproc) Cere Proc Ltd 10 th March 2008

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Paul Welham(paul@cereproc.com)CereProc Ltd10th March 2008 Company Confidential 2006

  2. Agenda • Purpose of today’s presentation • CereProc - Background • CereProc - What makes us unique • CereProc’s CereVoice and the need for character and emotion in synthetic voices • Summary Company Confidential 2006

  3. Purpose Of Today’s Presentation • To give you an overview of CereProc, our technology and what makes us unique and stand out in terms of solutions /services from other TTS vendors Company Confidential 2006

  4. CereProc– Who are we? • A spin out from the University Of Edinburgh • HQ in Scotland • Founded January 2006 by: • Dr Matthew Aylett (CTO) • Paul Welham (CEO) Our business is giving system integrators / developers the tools to enable them to make business solutions talk with a human voice Company Confidential 2006

  5. CereProcPartners • Closely tied with: • University of Edinburgh’s Centre For Speech Technology and Research (CSTR) • Universitat Pompeu Fabra Barcelona and Barcelona Media • Textic / Hidden Differences Group • ReadSpeaker • ScreenReader.net • + various other research institutions in Europe and USA Company Confidential 2006

  6. Why we started CereProc in 2005 • Speech synthesis just wasn’t that good that's why!! • We felt TTS offerings were old fashioned and technically conservative in their approach • The approach vendors had to speech synthesis had been monolithic • TTS vendors didn’t have a cost effective approach to exclusive “branded” voices • Change and Innovation was needed Company Confidential 2006

  7. CereProc - what makes us unique • CereProc is 100% an emotional synthetic speech company • CereProc only works with system integrators and responds to THEIR customers needs rapidly • With CereVoice we have the most flexible text to speech unit selection engine in the world • CereProc believe on going research is not only important, but necessary to give our customers competitive advantage and to improve quality – so we actively work with key academic institutions, to improve speech technology, the applications available and how the technology is deployed for all members of society Company Confidential 2006

  8. CereProc - what makes us unique • CereProc builds voices better and more cost-effectively than anyone else using our voice bulking and building technology • CereProc believe by using emotion we offer more characterful voices than anyone else • CereProc have voice cloning technology • CereProc can build voices from customer’s existing data Company Confidential 2006

  9. CereProc’s CereVoice provides system integrators today with • A range of custom voices varying in style and size, from small closed domain voices to full open domain, with emotion and character • An XML synthesis engine API which gives system integrators complete control of the engine’s front end – thereby allowing customisation and the ability to offer a fast response to their applications business requirements Company Confidential 2006

  10. CereProc’s CereVoice provides system integrators today with • We offer application developers: • An SDK which gives developers “refined” control of the back end performance e.g. customising latency • A plug in component based system across various operating system / environments on different platforms • CereVoice is resilient, scalable with a configurable core Company Confidential 2006

  11. CereProc’s our technical goals • To continue developing CereVoice to offer emotion, character and quality for your system integrators market needs • Speed up voice creation to under 3 hours per voice – making this process painless and cost effective • To launch a hybrid HMM and unit selection TTS system in Q3 2008 Company Confidential 2006

  12. CereProc’s Voices • Are More Than Just The Voice • They Are Expressive • Have emotional content • And can be fun Company Confidential 2006

  13. Why is emotion in a synthetic voice important? • It gives customers voices they can believe and trust in • Regional accents gives customers something they can relate too • It gives voices a “quality”, so that customers will want to listen to again and again • It makes it possible for customers to listen to synthetic speech for long periods of time • All of these elements will expand the number of applications available to system integrators to develop Company Confidential 2006

  14. Summary • Not just voices, but characters • We believe in continuous research &innovation • CereVoice from design has offered openness and flexibility • We offer rapid voice building and cloning • We help our partners get the best from our synthesis to meet their customers needs • CereProc’s founders have extensive experience of the synthetic speech business • But don’t just take my word for it Company Confidential 2006

  15. paul@cereproc.com If you are a systems integrator and would like to discuss working with CereProc – just drop me and Email Company Confidential 2006

More Related