Imagine if Siri understood every word you said, no matter how thick your accent, or how much you mumble.
Oxford researchers have developed a revolutionary new voice recognition technology, which is based on cutting-edge linguistics.
Rather than relying on statistical analysis, and running words through a database, the team at FlexSR worked with linguistics experts at the University of Oxford to develop a system based on the way the human brain identifies words.
This theory suggests that humans store a very basic acoustic representation of each word, made up of a few important distinctive characteristics.
When we hear a word we look out for these, and accept a wide variation in the sounds themselves, whether because of mumbling, or a thick accent.
The technology is easily adaptable to any spoken language, and because of its accuracy and the fact that it doesn’t have to be tailored to one voice, it is ideal for being used by lots of people at once, and isn’t bothered by background noise.
You say 'tomato', I say 'tomato', and FlexSR doesn’t mind either way!