摘要 |
PURPOSE:To correspond to the contents of speaking at a special subscriber side and to increase a translation efficiency by providing an automatic standard voice translating device at a telephone exchange. CONSTITUTION:The voice of an SUB-A1 is stored into an conversation voice spectrum storing part 6 for a specified caller. When the SUB-A1 starts the calling, a voice is inputted through a hybrid circuit part HYB4 to an interactive voice spectrum analyzing part 7 for a designated caller, for the analyzed spectrum, a comparing deciding voice identification 7 is executed with the contents of the storing part 6 and it is sent to a grammar correcting part 8 as voice pattern information. An error part such as a word order, an inflection and the termination of a word is corrected, sent to an insufficient vocabulary adding part 9, a correct word in the sequence relation of a sentence is added to voice pattern information and sent to a voice synthesizing part 11. The output of a standardizing device 3 is sent through an automatic standard voice automatic translating device 15 in an exchange 13 to a standardizing device 16 of an sub-B2. |