US 6,983,244 B2 | ||
Method and apparatus for improved speech recognition with supplementary information | ||
Jean-Claude Junqua, Santa Barbara, Calif. (US); Roland Kuhn, Santa Barbara, Calif. (US); Matteo Contolini, Santa Barbara, Calif. (US); and Rathinavelu Chengalvarayan, Santa Barbara, Calif. (US) | ||
Assigned to Matsushita Electric Industrial Co., Ltd., Osaka (Japan) | ||
Filed on Aug. 29, 2003, as Appl. No. 10/652,146. | ||
Prior Publication US 2005/0049860 A1, Mar. 03, 2005 | ||
Int. Cl. G10L 15/22 (2006.01) |
U.S. Cl. 704—231 | 27 Claims |
1. A method for improving recognition results of a speech recognizer, comprising:
receiving input speech from a user;
determining one or more candidate matches based on the input speech;
generating a prompt to the user for supplementary information by assessing ambiguity that remains after said step of determining
one or more candidate matches and tailoring said prompt to induce the user to supply additional information tending to resolve
said ambiguity;
receiving the supplementary data from the user as a response to said prompt; and
selecting one of the one or more candidate matches based on the input speech and the supplementary data.
|