The SpeechLab group is a part of the Institute of Information Technology and Electronics. It was founded in 1994 by Jan Nouza who has been involved in speech research for more than 20 years.
The SpeechLab's main research domain is speech recognition. In 1990s we were focusing our activities on solving the easier task: the recognition of isolated words and short phrases. Later our focus moved on the most challenging task - continuous speech recognition. Recently our own home-made systems are capable of classifying discrete words from large vocabularies (tens of thousands) in speaker-independent way, in real time (shorter than 1 s) with recognition rate higher than 95 %. As to the fluently spoken speech the rate we achieve is lower, about 90% for vocabularies with some 350 000 words. It should be noted that the continuous speech task is extremely complex in language like Czech, whose lexicon contains hundreds of thousands different words forms. Our policy is not to do theoretical research only but always we aim at verifying our methods, algorithms and results on practical applications. The best example of this effort are the systems: MobilDictate, ATT Toolkit, NewtonDictate, MyVoice or MyDictate.