| 摘要 |
PROBLEM TO BE SOLVED: To provide a voice monitoring system, etc., wherein the analyzed results of a vocalized voice such as an acoustic characteristics difference between the vocalized voice and the guidance voice, etc., are showed to the speaker so as to easily understand them. SOLUTION: A synthetic voice generating part 23 generates a synthetic voice for reading aloud a vocal text, and it drives headphones 11. When a speaker reads back it, the voice is comparted into parts representing respective characters by a voice recognition unit 222, then it is labelled and partitioned into respective frames. A frame analyzing part 223 obtains powers and fundamental frequencies of respective frames, and a voice information generating part 224 obtains powers and fundamental frequencies of voices representing respective characters. An indicator display 21 and a speaker display 13 display graphic representation, etc., of values of loudness and pitch of voices representing the respective characters, and they display the contents of the vocal text highlighting the currently vocalized part, then they display lines and letter strings in response to the operation of a mouse and keyboard of an input part 25. |
