摘要 |
Provided in embodiments of the present invention are a voice recognition method and terminal. The requirement of a user can be satisfied simply by sending an instruction once by the user without the assistance of a touch screen or inputting the instruction for multiple times. The method comprises: monitoring, by a voice wake-up device, voice information in a surrounding environment; if the voice wake-up device determines that the monitored voice information matches a voice wake-up model, then caching the monitored voice information having a first preset time period, and transmitting a triggering signal for triggering and activating a voice recognition device, the triggering signal being used to instruct the voice recognition device to read and perform a recognition, after being activated, on the voice information cached by the voice wake-up device; receiving, by the voice recognition device, the triggering signal transmitted by the voice wake-up device; and after the voice recognition device receives the triggering signal, the voice recognition device is activated and monitors second voice information having a second preset time period; and performing a recognition on the first voice information cached by the voice wake-up device and on the monitored second voice information, and obtaining a recognition result. |