Method and apparatus for processing speech information using a phoneme environment
AssigneeCanon Kabushiki Kaisha
InventorsToshiaki Fukada, Yasunori Ohora, Yasuhiro Komori, Takashi Aso
Life of this patent
Estimating the term…
Abstract
A speech information processing apparatus includes a statistical processing unit for extracting features by performing statistical processing of a feature file formed by extracting features of speech, such as the fundamental frequency and its variations, and the power and its variations of speech, from a speech file, and a label file in which a phoneme environment, comprising the accent type, the number of moras, the mora position, phonemes and the like, is considered, and a pitch pattern forming unit for forming a pitch pattern, in which phoneme environment is considered, based on the result of the statistical processing.
Claims23
A speech-information processing method comprising the steps of:
- performing statistical processing of stored speech data with respect to a phoneme environment parameter corresponding to a set of characteristics of a hierarchical configuration of sets of characteristics for each phoneme using a feature file formed by extracting features from the speech data, and a label file obtained by analyzing the speech data so as to be processed for each phoneme environment parameter;
- and
- forming a pitch pattern, in which the phoneme environment parameter is considered, based on the data subjected to the statistical processing.
A speech-information processing apparatus comprising:
- speech-file storage means for storing speech data;
- feature extraction means for forming a feature file by extracting features from the speech data stored in said speech file;
- analysis means for analyzing the speech data so as to be processed for each phoneme;
- and
- statistical processing means for performing statistical processing of the speech data stored in said speech file in consideration of a phoneme environment parameter corresponding to a set of characteristics of a hierarchical configuration of characteristic sets for each phoneme using data obtained from said feature extraction means and said analysis means;
- and
- pitch-pattern forming means for forming a pitch pattern, in which the phoneme environment parameter is considered, based on data obtained from said statistical processing means.
A computer usable medium having computer readable instructions stored therein for speech-information processing of stored speech data, the instructions comprising:
- a first set of instructions for causing statistical processing of the stored speech data with respect to a phoneme environment parameter corresponding to a set of characteristics of a hierarchical configuration of sets of characteristics of a phoneme using a feature file formed by extracting features from the stored speech data and a label file obtained by analyzing the stored speech data to be processed for each phoneme environment parameter;
- and
- a second set of instructions for causing formation of a pitch pattern considering the phoneme environment parameter and based on the data subjected to the statistical processing.
Dependencies are inferred from the claim text.