Phonetic recognition device independent unconnected with loudspeaker

Phonetic recognition device independent unconnected with loudspeaker

  • CN 101,334,997 A
  • Filed: 04/17/2002
  • Published: 12/31/2008
  • Est. Priority Date: 04/17/2001
  • Status: Active Application
First Claim
Patent Images

1. one kind is the method for speech recognition formation pronunciation model in telecommunication system, and described telecommunication system comprises at least one portable electric appts and server, and described method comprises:

  • In server, receive from electronic equipment (202,203;

    303,304) character string;

    In server, character string is changed (205;

         305) be at least one voice unit (VU) sequence;

    WithAt least one voice unit (VU) sequence is passed on (206 from server;

         306) to electronic equipment, it is characterized in that described voice unit (VU) sequence is the aligned phoneme sequence of text formatting,In the speech selection device of server, determine the language of character string, wherein, determine described language by the decision tree that character is specific, the specific decision tree of described character according to the character environment of each character represent at least a language probability andAccording to the language of in the speech selection device, determining, in server, character string is converted at least one aligned phoneme sequence of text formatting.

View all claims

    Thank you for your feedback