A kind of voice detection method based on multi-tone zone, related device and storage medium

Abstract

The present application discloses a voice detection method based on multi-tone zones, which is applied to the field of artificial intelligence. The voice detection method provided by the present application includes: acquiring the voice zone information corresponding to each voice zone in N voice zones; The sound area information corresponding to each sound area is used to generate the control signal corresponding to each sound area; using the control signal corresponding to each sound area, the voice input signal corresponding to each sound area is processed to obtain each sound area. The voice output signal corresponding to the sound zone; the voice detection result is generated according to the voice output signal corresponding to each voice zone. The present application also discloses a voice detection device and a storage medium. The present application can process speech signals from different directions in parallel based on multiple sound zones. In the scenario of multiple sound sources, the control signals can be used to retain or suppress speech signals in different directions, so that each user can be separated and enhanced in real time. voice, thereby improving the accuracy of voice detection.

Description

technical field [0001] The present application relates to the field of artificial intelligence, and in particular, to a method for detecting speech based on multi-tone regions, a related device and a storage medium. Background technique [0002] With the widespread application of far-field speech in people's daily life, in multi-sound source (or multi-user) scenarios, voice activity detection (VAD), separation, enhancement, and recognition are performed for each possible sound source. and call processing, has become a bottleneck for a variety of intelligent voice products to improve their voice interaction performance. [0003] In the traditional scheme, a monophonic pre-processing system based on the main speaker detection algorithm is designed. The pre-processing system generally adopts the method of azimuth angle estimation combined with signal strength estimation, or the method of azimuth angle estimation combined with spatial spectrum estimation. The speaker with the s...

AI Technical Summary

Problems solved by technology

[0004] However, when there are multiple speakers in the environment, judging the main speaker only by the arrival signal strength is flawed, because the main speaker may be farther away from the microphone array than the interfering speaker. Although the volume of the main speaker may be greater than that of the interfering speaker, the propagation loss of the speech signal in the space is greater, so the signal strength reaching the microphone array may be smaller instead, resulting in poorer effect on subsequent speech processing

Images