Hitachi, Ltd.

Akio Komatsu

Akira Ichikawa

Yoshiaki Asakawa

Nobuo Hataoka

Eiji Ohira

Kiyoshi Nagasawa

Kazuhiro Umemura

Yoshiaki Kitazume

Eiji Oohira

A voice coding system for separating and coding voice information into spectrum envelope information and voice source information, with the intention of compressing the amount of information for efficient coding of vocal audio signals through the control of the voice source information based on the fact that the spectrum envelope information and voice source information highly correlate with each other.

High efficiency voice coding system

A user interface system for information retrieval which includes apparatus for sequentially effecting syntactic and semantic analysis, every time a retrieval request sentence is inputted on words and clauses of the request sentence, apparatus inputted for creating partial syntactic and semantic expressions from successively obtained results of the syntactic and semantic analysis, apparatus for predicting words or clauses which are to be next inputted as the retrieval request sentence by making use of a semantic relation between the syntactic and semantic expression and an associated task, apparatus for displaying candidates for the predicted clauses, and apparatus for taking in selected clauses as subsequent clauses of the retrieval request sentence by and permitting a user to select at least one of the displayed candidates.

User interface system for permitting natural language interaction with

A speech recognition method makes it possible to improve the accuracy of recognition of input speech and is capable of operating on a real time basis. This is accomplished by generating from the input speech signal a difference signal which indicates whether the speech power of the input speech is increasing or decreasing for each frame. The similarity between the input speech and a standard pattern is then calculated for each frame, and this is then followed by correcting the similarity calculation on the basis of the generated difference signal and a difference signal relating to the standard pattern obtained from storage. The matching of the input speech and the standard pattern is then effected by using the corrected similarity, and the input speech is then recognized from the result of this matching. Thus, a spectrum matching distance weighted by power information of speech can be obtained in real time.

Speech recognition method

This speech signal recognition system compares the two-dimensionals pattern (time sequence of feature vectors) of an unknown signal to prestored standard references patterns for recognition, thus forming a corresponding two-dimensional comparison pattern of points of elemental Hamming distance differences. The sum of the pattern point distances is the similarity measure. To improve accuracy, partial patterns are selected (or 'masked') and tested sequentially, and the point values weighted relative to their location within the mask. The mask may be rectangular or oblique.

Continuous speech recognition method

Speech sound recognition is made using a reduced number of speech parameter elements, e.g., five correlation coefficients rather than sixteen spectral coefficients. The five correlation coefficients are derived from comparison of the spectral coefficients of unknown or standard sounds against the spectral coefficients of five highly-separable vowel-like sounds. Then, unknown-sound correlation coefficients are compared with standard-sound coefficients for recognition.

Growing community of inventors

Kodaira, Japan

Akio Komatsu