초록 ▼

In an apparatus for extracting information from an input speech signal, a preprocessor, a buffer, a segmenter, an acoustic classifier and a feature extractor are provided. The preprocessor generates formant related information for consecutive time frames of the input speech signal. This formant rela

In an apparatus for extracting information from an input speech signal, a preprocessor, a buffer, a segmenter, an acoustic classifier and a feature extractor are provided. The preprocessor generates formant related information for consecutive time frames of the input speech signal. This formant related information is fed into the buffer, which can store signals representative of a plurality of frames. The segmenter monitors the signals representative of the incoming frames and identifies segments in the input speech signal during which variations in the formant related information remain within prespecified limits. The acoustic classifier then determines classification information for each segment identified by the segmenter, based on acoustic classes found in training data. The feature estimator then determines, for each segment, the information required, based on the input speech signal during that segment, training data and the classification information determined by the acoustic classifier.

대표청구항 ▼

[ What I claim is:] [1.] An apparatus for determining, from an input speech signal, parameters defining the articulatory operation of a speech production system that generated the input speech signal, the apparatus comprising:dividing means operable for dividing the input speech signal into a succes

[ What I claim is:] [1.] An apparatus for determining, from an input speech signal, parameters defining the articulatory operation of a speech production system that generated the input speech signal, the apparatus comprising:dividing means operable for dividing the input speech signal into a succession of frames;a feature analyzer operable for determining acoustic features of the input speech signal during each frame;a segmenter operable for defining a succession of segments as a number of frames by (i) comparing signals representative of the acoustic features for a current frame with signals representative of the acoustic features of previous frames in a current segment; (ii) including he current frame in the current segment if said signals differ by less than a threshold value; and (iii) beginning a new segment with the current frame if said signals differ by more than said threshold value; anddetermining means operable for determining, for each segment, said articulatory speech production parameters using the acoustic features of the input speech signal during the segment and using stored reference data which relates said acoustic features to said articulatory speech production parameters.