| 主权项 |
1. A voice recognition method for an electronic apparatus, the voice recognition method comprising:
receiving, by the electronic apparatus, an input voice signal of a user; receiving, by the electronic apparatus, an image of the user; determining, by the electronic apparatus, characteristics of the user based on attributes of the input voice signal and attributes of the received image of the user; selecting, by the electronic apparatus, a voice model from a plurality of voice models stored in the electronic apparatus to determine the characteristics of the user based on the attributes of the input voice signal, the plurality of voice models comprising standardized attributes of respective average users and being associated with a plurality of characteristics of the respective average users; recognizing, by the electronic apparatus, the input voice signal using the selected voice model and identifying a user's command corresponding to the input voice signal of the user; and generating a responsive voice that is responsive to the recognized input voice signal based on the determined characteristics of the user and the selected voice model, wherein the determined characteristics of the user correspond to an age of the user, wherein a level of tone of the responsive voice has a level correlated with the age of the user corresponding to the determined characteristics, wherein the plurality of voice models are models of respective average users according to at least one among an age range and a gender, and wherein a database of the plurality of voice models is previously stored in the electronic apparatus with respect to the plurality of characteristics of the respective average users and the selecting the voice model comprises using the database of the plurality of voice models. |
