The Organisation for Economic Co-operation and Development (OECD) is conducting a study to understand the opportunities and challenges of using artificial intelligence (AI) for accessibility, and in particular to facilitate people with disabilities’ access to the labour market. The end-goal of the study is to form public policy recommendations with regards to promoting innovation in AI that can improve labour market accessibility. Policy makers from the UK as well as other international countries will be included in these conversations.
A final report will be launched by the OECD in 2023. Previous OECD reports on AI have concluded that in net terms, the emergence of AI technology hasn’t thus far reduced the number of jobs in the labour market, but rather it changes the terms of some jobs (both low-skilled and highly-skilled roles), while creating new opportunities for the future. [See article “The impact of AI on the labour market: is this time different?” on the OECD website here.]
The research team, led by Chloé Touzet, approached Liopa to request our participation in the study, to gain a better understanding of how the SRAVI app can assist people who are vocally-impaired.
Liopa was one of a number of innovative firms approached, because of our role in developing AI-based solutions for different aspects of accessibility. Our views on the opportunities offered by AI, as well as the challenges in the field, have been shared for the report.
The research complements the OECD’s AI-WIPS (AI in Work, Innovation, Productivity and Skills) research programme, centred on the impact of AI on the labour market, skills and social policy. The OECD can be found on Twitter at @OECD_Social.
How the SRAVI app uses AI to help people with AAC needs or vocal impairments
Liopa was incorporated in 2015 to commercialise more than 15 years of PhD research at QUB in the field of speech and image processing. Using this deep knowledge, we developed an automated lipreading solution. Our technology – rooted in machine learning and AI – can decipher what a person is saying from lip movements alone.
Liopa is the first company to release commercially available VSR (Visual Speech Recognition) technology, with the first application being our SRAVI app (www.sravi.ai).
SRAVI is a simple lipreading application that requires just a smartphone. It brings a voice to the voiceless.
With SRAVI, we help patients who have lost the ability to speak, to communicate with their loved ones & carers. SRAVI is deployed onto a smartphone, and when pointed towards the patient, the app analyses their lip movements to decipher what they are trying to say. SRAVI is currently in use in Intensive Care units within the NHS – our first commercial customer being Lancashire Teaching Hospitals NHS Trust.
Liopa’s lipreading technology can also be used in a range of other applications, including keyword spotting, active speaker detection, and silent communication.
Liopa has received financial awards from Innovate UK, DASA and the Future Fund.
What’s unique about SRAVI?
Traditional communications devices for the speech impaired (e.g. GRID-based applications) are unsuited for patients in Critical Care environments. To use these complex systems, patients need to be relatively well. These devices require a very active degree of interaction from the users – training is required, and also levels of mental lucidity & manual dexterity uncommon in patients in Critical Care. In contrast, SRAVI only requires that the patient mime what they are trying to say, and is therefore a more natural solution in these environments.
In contrast, SRAVI is “ultra-easy” – and here’s why:
- Users can be up and running in seconds
- You simply aim a smartphone camera at the person, and it reads their lips from a defined set of phrases
- They can communicate with family members, doctors and nurses instantly, with basic but essential phrases like “I’m hungry. I need the toilet.”
ICU Consultant Dr Shondipon Laha of the Lancashire Teaching Hospitals NHS Trust explained SRAVI’s utility:
“Not being able to speak is very frustrating for the patient, and for staff. We can deal with patients’ needs much more rapidly and the rehabilitation process is much smoother if we can improve communication. SRAVI is absolutely fundamental to the successful treatment of extremely ill patients.”
SRAVI has helped patient Nathan Armstrong. Nathan said “I love how simple SRAVI is to use. SRAVI enables me to tell people what I need.”
