To provide a hypertext voice control method, device and program for facilitating the description of the conversion of hypertext into voice, while reducing Proxy loads, increasing the range of voice recognition, and making it possible to confirm voice recognition results through voice synthesis.
To relay hypertext, an instruction part is extracted and equipped with an associated text part, operation instruction information for providing instructions on the operation of a voice recognition part, and a instruction indicating that an instruction corresponding to the recognized language for voice recognition is to be executed. The text part is analyzed to derive a phonetic representation and specify the language for voice recognition. The instruction corresponding to the language for voice recognition is executed according to the user's voice.
AMAKASU TETSUO