Method, apparatus and device for optimizing wakeup model, and storage medium

Abstract

Embodiments of the invention provide a method, an apparatus and a device for optimizing a wakeup model, and a storage medium. The method comprises the following steps: acquiring a training set and a verification set; performing iterative training on the wakeup model according to the training set and the verification set; in the iterative training process, periodically updating the training set andthe verification set according to the wakeup model and a preset corpus, and continuing to perform iterative training on the wakeup model according to the updated training set and verification set; and outputting the wakeup model when a preset stopping condition is met. According to the method, the training set and the verification set are updated periodically according to the wakeup model and thepreset corpus in the iteration process, so that the optimization efficiency and effect of the wakeup model can be improved, the stability and adaptability of the wakeup model can be improved, and theoverfitting is avoided.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to an optimization method, device, equipment and storage medium of a wake-up model. Background technique [0002] With the rapid popularization of mobile smart devices, voice, which is a natural and convenient way for humans, has become increasingly popular in human-computer interaction. At present, smart devices with voice interaction functions such as mobile phone voice assistants, smart speakers, voice dialogue robots, and voice-controlled TVs are emerging in an endless stream. Major Internet companies and innovative companies have poured into this hot battlefield. According to user feedback and user research, voice interaction has indeed greatly improved the convenience for users. [0003] As an important part of voice-related technology, Hotword Detection technology is in the first link in the entire voice interaction, and plays a decisive role in the overall in...

AI Technical Summary

Problems solved by technology

However, due to the limited number of false wake-up data in the tuning set, the overall optimization stage is slow; and as the wake-up model is optimized, the revenue brought by the pre-collected tuning set drops rapidly, resulting in a lower optimization efficiency and effect on the wake-up model. Poor, affecting the stability of the model

Images