Semantic comprehension model training method, semantic comprehension method and device and storage medium

Abstract

The invention provides a semantic comprehension model training method. The semantic comprehension model training method comprises the following steps: recalling a training sample matching a vehicle-mounted environment in a data source; carrying out boundary corpus expansion processing on a statement sample with noise matching the vehicle-mounted environment; annotating statement samples with noise, which are subjected to boundary corpus expansion processing and matched with the vehicle-mounted environment, so as to form a first training sample set; processing the second training sample set through a semantic comprehension model; and according to the update parameters of the semantic comprehension model, performing iterative update on the semantic representation layer network parameters andthe task-related output layer network parameters of the semantic comprehension model through the second training sample set. The invention further provides a semantic comprehension method and deviceand a storage medium. According to the semantic comprehension model training method, the training precision and the training speed of the semantic comprehension model can be improved, so that the semantic comprehension model can adapt to a vehicle-mounted environment full duplex use scene, and the influence of environmental noise on the semantic comprehension model is avoided.

Description

technical field [0001] The invention relates to machine learning technology, in particular to a semantic understanding model training method, a semantic understanding method, a device and a storage medium. Background technique [0002] In the usage scenario of full-duplex voice interaction, it is necessary to implement the following operations in a multi-sound source environment where multiple sound sources continuously emit sound at the same time: for example, the recognition of voice identities (male, female, and child), triggering dialogues with different contents, voice Emotion recognition, music / singing recognition, etc.; environmental processing, background noise recognition and echo cancellation, in this process, the semantic understanding model has nothing to do with the field of background noise and other people's chatting in the full-duplex dialogue scene (OOD, Out -Of-Domain) corpus is easier to be listened to by assistants. If such corpus is misresponsed by smart...

AI Technical Summary

Problems solved by technology

[0002] In the usage scenario of full-duplex voice interaction, it is necessary to implement the following operations in a multi-sound source environment where multiple sound sources continuously emit sound at the same time: for example, the recognition of voice identities (male, female, and child), triggering dialogues with different contents, voice Emotion recognition, music / singing recognition, etc.; environmental processing, background noise recognition and echo cancellation, in this process, the semantic understanding model has nothing to do with the field of background noise and other people's chatting in the full-duplex dialogue scene (OOD, Out -Of-Domain) corpus is easier to be listened to by assistants. If such corpus is misresponsed by smart assistants, the interaction success rate is low and affects the user experience

Images