Differential-training-based voice wake-up customization optimization method and system

Abstract

The invention relates to a differential-training-based voice wake-up customization optimization method and system. After acoustic data with marks are given, neural network forward propagation is carried out by combining an initial neural-network-based acoustic model; a modeling probability for wake-up word marking is obtained by combining a marking sequence and a frame-by-frame acoustic probability; statistics is carried out on lots of texts to obtain a phoneme-level language model and a corresponding searching network is constructed; wake-up word competition item modeling is carried out based on the searching space of the phoneme-level language model and a frame-by-frame acoustic model, thereby obtaining a posterior probability; differential training of the acoustic model is carried out by combining the wake-up word competition item modeling and the wake-up word marking modeling; and the acoustic model obtained by the differential training determines whether to carry out a wake-up operation by comparison of a wake-up word confidence coefficient with an automatic estimation threshold value.

Description

technical field [0001] The present invention relates to a technology in the field of voice recognition, in particular to a method and system for optimizing customized voice wake-up based on discriminative training. Background technique [0002] The task of the customizable voice wake-up (Voice Wake-up) system is to automatically discover and locate some pre-specified command words (wake-up words) from a continuous speech. The customizable feature is reflected in the fact that the wake-up word detection model does not depend on the user-specified wake-up word, so that the user's wake-up word can be easily changed without modifying the model. The main voice wake-up techniques include the early dynamic time warping method, the current hidden Markov model-based method, and the deep learning-based method. Contents of the invention [0003] The present invention aims at the defects and deficiencies such as the inability to clearly distinguish between wake-up words and non-wake-...

AI Technical Summary

Problems solved by technology

[0003] The present invention aims at the defects and deficiencies such as the inability to clearly distinguish between wake-up words and non-wake-up words in the prior art, or that can only be realized by customizing the speech model, thereby increasing the implementation cost and reducing the robustness, etc., and proposes a The customized voice wake-up optimization method and system based on discriminative training uses context-independent phoneme acoustic decoding to perform efficient search, thereby completing the confidence selection of wake-up words

Images