摘要 |
PROBLEM TO BE SOLVED: To provide a device for learning pronunciations of various regions. SOLUTION: The electronic dictionary 1 has a CPU6 to generate voice composited data from the text by correlating each phoneme in the English text with the stress data of the standard English, a local voice parameter storing table 87 to store the stress correction factors to the stress data of the standard English for each kind of local English, and an voice output 3. The CPU6 specifies a certain kind of local English stored in the local voice parameter storing table 87 as an English to learn and corrects the stress data in the voice composited data by using the stress correction factor for the English to learn. The voice output 3 outputs the phonemes in the above text with the stress of the English to learn, based on the voice composited data after corrected in the CPU6. COPYRIGHT: (C)2009,JPO&INPIT
|