VOICE INPUT JAPANESE LANGUAGE SENTENCE PREPARING DEVICE

• Main Authors: TAKAGI HIDEYUKI, KUSUHARA HISAYO, NAKAJIMA AKIKO
• Format: Patent
• Language: English
• Published: 06.06.1986
• Subjects: ACOUSTICS; CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

PURPOSE:to improve the operability by providing a recognition candidate example buffer for storing plural pieces of a higher rank of a recognition result, as candidate examples. CONSTITUTION:A voice recognizing part 10 recognizes a voice of a Japanese language, for instance, by a monosyllable unit and stores a result in a recognition result buffer 11. A syllable candidate example buffer 12 stores plural pieces of a higher rank of the recognition result buffer 11, as candidate example monosyllables, and also displays them on a syllable candidate example display part 13. a candidate example string preparing part 14 prepares a candidate example string by selecting one of the candidate example monosyllables stored in the syllable candidate example buffer 12, at every monosyllable of an input voice, and stores it in a candidate example string buffer 15. A language processing part 16 extracts a candidate example string having a semantics as a Japanese language, in the candidate examples stored in the candidate example string buffer 15. A Chinese character converting part 18 converts a candidate example string extracted by the language processing part 16 to a Chinese character and KANA (Japanese syllabary) mixed notation, and displays it on an output display part 19. By depressing a monosyllable next candidate key 20, the syllable candidate example buffer 12 is rewritten in accordance with the contents of the recognition result buffer 11.