Concept‐to‐speech conversion for reply speech generation in a spoken dialogue system for road guidance and its prosodic control Free

Because of advancements of speech and language processing, a number of spoken dialogue systems have been constructed. However, because most of them adopt existing text‐to‐speech synthesizers to generate output speech, it is rather difficult to reflect all the linguistic information that is obtained during the reply sentence generation. To resolve this situation, a framework must correctly reflect higher‐level linguistic information, such as syntactic structure and discourse information, on the prosody of output speech: concept‐to‐speech conversion, where reply sentences are generated from information (to be transmitted) and converted into speech in a unified process. A spoken dialogue system for road guidance was constructed, and concept‐to‐speech conversion was realized in the system. The linguistic information of the generated sentence is handled in tag LISP form to retain the syntactic structures throughout the process. Moreover, a new method of sentence generation from concept was realized with this system: it handles a concept in phrase units and aggregates them to form a sentence. It is tested whether the linguistic information could be reflected properly on the prosody of output speech. Results of listening experiments verified the effectiveness of our proposed method.