| 摘要 |
PROBLEM TO BE SOLVED: To generate association information for synchronously reproducing text data and voice data with high precision very easily, in a text and voice synchronizing device. SOLUTION: Through vocal input of a language learning sentence with a model spoken voice, a recognized voice waveform and a recognized text are displayed in an input voice waveform display area 45 and a recognized text display area 44 in order while made to correspond to each other as the spoken voice is input; and the speech-recognized text can easily be acquired as a learning text corresponding to a spoken voice of the language learning sentence and text and voice synchronization data with the elapsed time tn of the vocal input can easily be generated and obtained by speech-recognize words wn. On the basis of the text and voice synchronization data 12e, the learning text is sequentially identified and displayed according to speaking speeds (time) by words and a time code file for synchronous reproduction for synchronously reproducing the spoken voice can be generated very easily. COPYRIGHT: (C)2005,JPO&NCIPI
|
