Voice wake-up method and device based on artificial intelligence, and computer equipment

Abstract

The invention provides a voice wake-up method and device based on artificial intelligence, and computer equipment. The method comprises steps that windowing enframing operation of online recorded voice is carried out to acquire at least one voice frame; characteristic extraction of the voice frame is carried out to acquire voice characteristics; a voice wake-up model based on the convolutional neural network is utilized to carry out calculation of the static voice characteristics contained in the voice frame, and posterior probabilities of the static voice characteristics contained in the voice frame corresponding to a non-wake-up word class and a wake-up word class are acquired; if the posterior probability of the static voice characteristics contained in the voice frame corresponding to the wake-up word class is greater than a pre-determined threshold, the online recorded voice is determined to contain wake-up words. The method is advantaged in that parameter quantities of the voice wake-up model based on the convolutional neural model can be effectively reduced, so the operand reduction effect is realized, and the voice wake-up effect can be improved.

Description

technical field [0001] The present application relates to the technical field of voice recognition, in particular to an artificial intelligence-based voice wake-up method, device and computer equipment. Background technique [0002] In smart home or voice interaction systems, voice wake-up (wakeword) technology is widely used. However, the effect and calculation of voice wake-up greatly reduce the actual application experience of voice wake-up, and increase the requirements of voice wake-up on device hardware. For example, if the false wake-up rate of voice wake-up is greater than a threshold in the application, such as 1 / 3 Such a frequency of false wake-up triggers may arouse user disgust; on the other hand, if the computing power of voice wake-up exceeds the computing power of some low-end chips, it will restrict the use of voice wake-up technology in many products. [0003] In the existing related technologies, the technology used for voice wake-up is the keyword-spottin...

AI Technical Summary

Problems solved by technology

[0004] However, the voice wake-up technology of the above keyword recognition method has a large amount of model parameters, and for different wake-up words, the design of the filler (filler) must be changed, and the corresponding decoding parameters and detected tricks must be adjusted. It is difficult to have a unified algorithm to ensure that the effect of each wake-up word is at a stable level, and after the filler is set, the misrecognition level of the wake-up word is fixed, and it is not possible to flexibly and conveniently correct the misrecognition Wake words for tuning and learning

Images