Projects


We are currently working on the following projects:

Voice Input Voice Output Communication Aid 2 (VIVOCA2)

The VIVOCA2 Project is funded by the National Institute for Health Research (NIHR) New and Emerging Applications of Technology (NEAT) funding stream for 2 years from June 2010. The project partners are: the University of Sheffield School of Health and Related Research (ScHaRR) Rehabilitation and Assistive Technology group and the Speech and Hearing group in the Department of Computer Science; Barnsley Hospital Assistive Technology team; Toby Churchill Ltd.; Elpedium and Medipex Ltd. The VIVOCA2 project is a development of the original PDA-based proof of concept device developed under the original VIVOCA project which was the first time such a concept has been successfully tested with users. More information is available on the blog.

MyVOCA

Funded by the University of Sheffield for 3 months from January - March 2011, this project aims to continue voice banking work, building personalised synthetic voices for people with severe speech impairment including degenerative conditions such as Motor Neurone Disease.

Speech-driven Environmental Control Systems (SPECS)

Funded by the Department of Health HTD programme from Feb 2006 for three years, in collaboration with Toby Churchill Ltd., XoVox Communications Ltd., and Medipex Ltd. SPECS aims to develop a novel environmental control system for disabled and elderly people, controlled by automatic speech recognition.

REdRESS (Recognition and Reconstruction of Speech following Laryngectomy)

Funded by Action Medical Research in collaboration with The University of Hull, Hull Royal Infirmary and The Digital Design Studio (GSA). REdRESS aims to develop a computer based system capable of detecting movement in the vocal apparatus which allows the reconstruction of speech without acoustic signals. The potential benefit is to provide improved speech quality output and to be less debilitating for patients than existing methods of voice restoration.

Previous projects

Previous projects that we have worked on include:

Voice Input Voice Output Communication Aid (VIVOCA)

Funded by the Department of Health NEAT Programme from Jan 2005 for three years, VIVOCA aimed developed a portable speech-in/speech-out communication aid for people with disordered or unintelligible speech, specifically people with moderate to severe dysarthria.

Ortho-Logo-Paedia (OLP)

Funded by the EU Framework 5 Quality of Life Programme from 2002 to 2004. Developed a software toolkit to help speech therapists by providing feedback based on automatic speech recognition and on ‘kinematic maps’ which provide visual feedback converting instantaneous acoustics onto position on a trainable, two dimensional display.

Speech Recognition for People with Severe Dysarthria (STARDUST)

Funded by the Department of Health NEAT programme from 2000 to 2003, we showed that small-vocabulary speech recognition can function well for people with severe speech disorders, with sparse training data. Introduced measures of consistency of articulation and provided clients with visual feedback to practice improvement in consistency. Field trials showed that disabled clients were able to give commands to assistive technology by voice twice as quickly as with their normal interface.