| 摘要 |
The present invention relates to a speech recognition method in an apparatus for users which includes a step (a) of extracting a speech propriety for identifying an user from a speech received from the user, a step (b) of searching the extracted speech property in the speech property stored in a device, a step (C) of determining a text corresponding to the speech of the user if the extracted speech property exists. By using the present invention, more accurate speech recognition can be carried out at low cost in the apparatus of a multiplex agency. The apparatus of the multiplex agency can be conveniently controlled. [Reference numerals] (S100) Register voice features; (S200) Control by using the voice features; (S300) Correct the voice features |
