About the project
VocaTempo, working as an app on a tablet device, will recognise the speech of children with severe speech impairments. The app will speak out a clear version of the message the children wish to convey, thus allowing them to communicate more effectively.
Children can – with the assistance of a parent, teacher or speech and language therapist – train the app to recognise the manner in which they speak certain vocalisations (spoken words). Once the app is trained to recognise a user's voice, it can be controlled vocally. Users will be able to navigate through the app using their speech, and will also be able to activate text-to-speech cells, which will read a pre-programmed message using synthetic speech.
It is a hands-free method of communication, allowing users with dysarthria to communicate with everyone quickly, easily and without breaking eye-contact with their communication partner. A grid system, which can be personalised, allows users to move through the app with minimal effort, and new vocalisations can be added at any time.
The Small Business Research Initiative (SBRI) granted funds to assist with the phase one development. This has allowed Therapy Box® to develop a proof of concept application, which is being tested with real users. Further development based on user feedback is expected later in the year, prior to a wider release.
A team led by Professor Mark Hawley from CATCH created the advanced speech recognition software, specifically engineered to work with dysarthric voices. Barnsley Hospital provided assistance and support in the conception of the software and in testing the recognition module with real users. CATCH will continue to refine their software throughout the VocaTempo project, and a team at Barnsley Hospital, led by Simon Judge will be working with young people in focus groups to provide the feedback needed to make a truly valuable and revolutionary AAC product.