Speech signal processing method, apparatus and electronic device

Abstract

The embodiments of the present invention disclose a speech signal processing method, a speech signal processing apparatus and an electronic device. The method comprises the following steps that: when the electronic device is under a sleep state, speech signals are received, whether an interaction instruction corresponding to the received speech signals is a wake-up instruction is judged; if the interaction instruction corresponding to the received speech signals is a wake-up instruction, the sleep state is switched to a working state, and the sound source direction of the received speech signals is positioned as a user sound source direction; speech signals are further received, noise suppression processing is performed on speech signals in the further received speech signals, wherein the speech signals in the further received speech signals do not come from the user sound source direction, so that user speech signals are obtained; and an interaction instruction corresponding to the user speech signals is responded. According to the speech signal processing method, the speech signal processing apparatus and the electronic device of the invention, the electronic device performs noise suppression processing on the speech signals in the further received speech signals which do not come from the user sound source direction, so that the obtained user speech signals are speech signals emitted by a user in the user sound source direction, and therefore, a correct response can be made, and user experience can be improved.

Description

technical field [0001] The invention relates to the technical field of voice signal processing, in particular to a voice signal processing method, device and electronic equipment. Background technique [0002] At present, there are more and more products on the market with voice interaction functions, such as electronic devices such as smart speakers and robots. After receiving the wake-up command, these electronic devices switch from the standby state to the working state, and receive the voice signal through the microphone array, that is, pick up the sound, and then recognize and analyze the voice signal, so as to respond to the corresponding interaction of the voice signal instruction. [0003] After receiving the wake-up command, the above-mentioned electronic device with voice interaction function receives voice signals from various sound sources in the surrounding environment through the microphone array, and recognizes the sound source orientation corresponding to th...

AI Technical Summary

Problems solved by technology

[0004] Under normal circumstances, the application of the above method can better process the voice signal, but if there are one or more sound objects around the user whose volume is higher than the user's volume, the electronic device with voice interaction function will receive the voice signal The direction of the sound source corresponding to the one with the loudest volume is identified as the direction of the user’s sound source, and the voice signal with the loudest volume is identified and analyzed to obtain an interactive command, and then an incorrect response will be made, resulting in a poor user experience

Images