摘要 |
PROBLEM TO BE SOLVED: To provide a voice simultaneous processor capable of dividing a voice recognition result into appropriate processing units.SOLUTION: The voice simultaneous processor includes: an acquisition section; a voice recognition section; a detection section; and an output section. The acquisition section acquires a voice signal. The voice recognition section generates a determined character string which is a character string corresponding to a voice period in which the voice recognition processing has been made on the voice signal and converted into a character string, and one or more candidate character strings corresponding to a voice period which is a voice period following the determined character string which is under processing of voice recognition and character string as a conversion result is not determined. The detection section detects a character string as a processing unit character string, when one or more combination character strings, in which the determined character string and the one or more candidate character strings are coupled, are divided based on a division point which represents a position of a morpheme as a start position of a processing unit in natural language processing, and when any character string common to each of the combination character strings is included in a determined character string. The output section outputs the processing unit character string. |