PDAs and Mobile Phones with Humanlike Speech Recognition
Published: Jun 20, 2007Researchers at the Russian Academy of Sciences created speech recognition software that needs little processing speed, processor time and memory consumption and is suitable for use in mobile devices, such as PDAs and mobile phones.
Researcher Vyacheslav Anciperov and colleagues from the Institute of Radio Engineering and Electronics of the Russian Academy of Sciences have developed a system that can understand speech like a human. The system collects data from the word’s meaning regardless of how it is spoken and it ignores components not related to the meaning, which leads to more accurate speech recognition.
Humans understand various tones, volumes and intonations by separating these from the spoken words. Speech applications until now make comparisons between data previous collected and the speech input and then try to make a match.
Anciperov and colleagues say in an article of Betterhumans, that they have produced a prototype software interface that works differently, singling out essential elements of speech using a hierarchical structure. “Like in a musical composition, one can recognize more or less high levels of organization—rhythm, main theme, arrangement, so we have also learned to single out the ranges in the speech flow which carry the major semantic loading,” says Anciperov. “The approach simplifies tasks for computers,” he says, “and could therefore allow for better speech recognition on less powerful computers such as those in mobile phones and PDAs. Our system wins in processing speed and in processor time and memory consumption as compared to those of all known similar systems.”
Source: Betterhumans
