Belfast start-up Liopa secures $1m funding for lip read technology
Technology has multiple uses including improving quality of voice control systems like Siri or Alexa
Richard McConnell and Liam McQuillan from Liopa with Hal Wilson, Partner, techstart NI.
Liopa, a Belfast-based start-up that has developed an innovative lip-reading technology, has secured $1 million in funding.
Developed following 10 years of research by the Centre for Secure Information Technologies (CSIT) at Queen’s University, LipRead uses the camera on a smartphone or tablet to analyse lip movement and decipher words and phrases.
LipRead’s primary function will be improving the accuracy of commonly used voice-driven assistants (such as Siri, Cortana or Alexa) and voice activated applications in noisy environments such as vehicles, offices & factories.
Other uses include as an aid for people with hearing impairment, for better human/robotic interaction and as a potential tool to prevent fraud by adding an extra password layer to programmes.
The funding will help firm bring its Visual Speech Recognition (VSR) platform to market. Led by its initial backers, Techstart NI, and QUBIS, its latest investors include the AI Seed Fund which exclusively supports start-ups that harness Artificial Intelligence to create leading edge tech solutions. The funding round was also supported by Co-Fund NI which is managed by Clarendon Fund Managers.
The firm is led by entrepreneurs Liam McQuillan and Richard McConnell. Dr Darryl Stewart and Dr Fabian Campbell-West, world-renowned authorities in Visual Speech Recognition & Computer Vision, are also part of the Liopa team.
According to Liam McQuillan: “Voice is forecast to overtake the keyboard and mouse as the primary means of interacting with mobile & computing applications. The big stumbling block however is the accuracy of current voice technologies in real world environments - commands get misinterpreted frequently due to background noise. LipRead will augment current technologies and provide this much needed accuracy, thus ensuring the success of the ‘Voice UI’.