摘要 |
PURPOSE:To provide an automatic interpretation system capable of dealing with various situation occurring in process of usage, such as the occurrence of an erroneous voice recognition, erroneous translation, etc. CONSTITUTION:This system is provided with plural input/output means provided with voice input parts 3 and 7 for inputting conversation, instruction input parts 2 and 6 for inputting instruction information, voice output parts 4 and 8, and display parts 1 and 5 and added with attributes showing the languages of mutually different kinds. The system is also provided with a voice recognition means 10 voice-recognizing a speech from the voice input part included in one input/output means based on the attribute and coding it, an automatic bi- directional translation means 12 translating the code into a code corresponding to a language with another attribute, a voice generation means 11 converting the code into voice and a conversation means 13 conversing with the operator of the one input/output means, namely a speaking person, for confirming a result when the recognition result or the translation result of the speech can not be specified and conversing about the conversation with the operator of the other input/output means, namely a hearing person, during conversation. |