Sound source positioning method and system, electronic equipment and storage medium

Abstract

The invention discloses a sound source positioning method and system, electronic equipment and a storage medium. The sound source positioning method comprises the steps of dividing a voice signal received by a reference microphone into at least one sub-band, wherein the reference microphone is one microphone in a microphone array; if voice exists in the voice signal through VAD detection, for eachsub-band, calculating the voice existence probability according to the likelihood ratio of voice relative noise obtained in the VAD detection process; and calculating the maximum controllable response power of the microphone array by using an SRP-PHAT algorithm based on sub-band voice existence probability weighting, and determining the direction of the sound source according to the angle corresponding to the maximum controllable response power. According to the invention, the voice existence probability of each frequency band is calculated by utilizing the intermediate result of the VAD detection voice and is weighted into the original SRP-PHAT algorithm, so that the sound source positioning performance under the low noise ratio is improved under the condition of not increasing the additional calculation amount.

Description

technical field [0001] The invention relates to the field of signal processing, in particular to a sound source localization method and system, electronic equipment and a storage medium. Background technique [0002] At present, speech recognition performs better in scenarios with high signal-to-noise ratio, no reverberation, and relatively close sound source distance. However, as the voice interaction scene expands to applications such as smart speakers, cars, and robots, the single-microphone system cannot meet practical applications, and the microphone array can pick up sound sources in different directions to achieve space-time signal processing to improve the performance of far-field voice interaction. been widely studied. [0003] Sound source localization is the first step in the signal processing of the microphone array and is implemented during the voice wake-up phase. Its accuracy will directly affect the effect of subsequent signal processing, usually combined w...

AI Technical Summary

Problems solved by technology

However, the performance of the algorithm will drop sharply in low signal-to-noise ratio and reverberant environments, and its large amount of calculation will reduce the practicability

[0005] At present, the improvement of the SRP-PHAT algorithm is concentrated in two aspects: one is to improve the efficiency of the algorithm by optimizing the search strategy, but the results are not good in the case of noise, reverberation, and interference; the other is to calculate the time and frequency of each time-frequency point The "energy" envelope state of the direct wave determines the direct wave. Although the performance in high reverberation and interference environments is improved, it increases the amount of computation of the original SRP-PHAT algorithm and reduces the practicability of the algorithm.

Images