Now almost 18 months into the Nuvoic Project we’re starting to see some great outcomes, both in terms of participants using the Voiceitt app to do more things more independently, and also some new developments in the user interface and app performance, in response to participants’ feedback.
The project is led by developers Voiceitt and aims to improve access to voice recognition technologies for people who have dysarthric speech. The Karten Network team leads on user testing, supporting people with non-standard speech to try the app for support with communication or Smart Home control, and give feedback on what works well and what could be improved.
We now have over 60 participants enrolled in the project, and we’re delighted to welcome new participants from Leonard Cheshire, Leuchie House in Scotland and Linkage College who join those already involved from Beaumont College, Caritas St. Joseph’s, Cedar Foundation, Enable Ireland, FitzRoy, The Grange Centre, Homefield College, National Star, New Bridge Horizons, The Oaks College, Portland College, SeeAbility, St John’s College and Young Epilepsy.
As Covid-19 restrictions have eased, more participants have been using Voiceitt to support communication out-and-about: to socialise, order food and drink in a café or to use public transport, for example. (Our team has also been able to get out and visit participants in person, which has been great!) We also have participants trying out a range of different Smart Home controls. Some use voice control to play music, radio or audiobooks via a smart speaker, ask about the news, weather or sports results, or tell jokes. Others are using Voiceitt to control other devices like smart bulbs, plugs, windows, blinds and TVs. Check out our participant videos to see some examples for yourself!
We keep in touch with participants throughout their testing and pass all feedback on to the developers. We’ve already seen improvements in response to user feedback: it’s now quicker and easier to train new commands, and changes to the user interface to make the app easier to navigate and more fun to use, for example. Switch access and predictive text have also been implemented. Thanks so much to everyone involved for all your fantastic feedback so far, please keep it coming!
Over the next few months, as well as continuing with testing and feedback, we’ll also be starting to recruit participants to donate voice samples through Voiceitt’s ‘Ensemble’ scheme. Voiceitt want to move from ‘discrete recognition’, where users train the app in advance to recognise how they say each phrase or command, to ‘continuous recognition’ of spontaneous speech, and need to collect a lot of new voice data to support this work. While this won’t benefit participants immediately, we can offer vouchers or equipment to participants and support teams who take part. Details to follow soon – watch this space!
Get in touch!
We would love to hear from you if you, or someone you know may be interested in taking part in testing and/or voice sampling. Please email our project co-ordinator: firstname.lastname@example.org, or you can find more information and get in touch via our project web pages.
Article meta data
Clicking on any of the links in this section will take you to other articles that have been tagged in the same category.