| CPC G10L 15/26 (2013.01) [G10L 15/063 (2013.01); G10L 25/24 (2013.01)] | 18 Claims |
|
1. A speech instruction recognition method, comprising:
acquiring a target speech;
processing the target speech to obtain a target speech vector corresponding to the target speech;
performing speech recognition on the target speech to obtain a target speech text of the target speech, and processing the target speech text to obtain a target text vector corresponding to the target speech text;
generating a to-be-trained instruction recognition model;
obtaining a pre-trained instruction recognition model by performing an iterative training on the to-be-trained instruction recognition model with sample speeches;
inputting the target speech vector and the target text vector to the pre-trained instruction recognition model to obtain an instruction category corresponding to the target speech, so that a corresponding operation is performed according to the obtained instruction category,
wherein the step of inputting the target speech vector and the target text vector to the pre-trained instruction recognition model to obtain an instruction category corresponding to the target speech comprises:
performing concat on the target speech vector and the target text vector to obtain a concat vector; and
inputting the concat vector to the pre-trained instruction recognition model to obtain the instruction category corresponding to the target speech;
performing a response or operation using a smart device or an IoT (Internet of Things) system according to the instruction category.
|