Method and device for tracking speaker by applying microphone, and computing equipment

Abstract

An embodiment of the invention relates to the technical field of communication, and discloses a method and a device for tracking a speaker by using a microphone, and computing equipment. The method comprises the steps of: recognizing the speaker through sonar imaging; acquiring mouth coordinates of the speaker and a distance from the microphone to the mouth coordinates; and controlling the microphone to directly face the mouth coordinates of the speaker in a first preset range in real time by a mechanical arm and dynamically track the speaker. According to the method and the device of the invention, the accurate position of the speaker can be obtained from a sonar, the development of a microphone front-end algorithm to be more centralized in the development of a certain amplitude angle through the dynamic adjustment of an angle between the microphone and the speaker, can save the development time, and achieves a better voice effect.

Description

Technical field [0001] The embodiments of the present invention relate to the field of communication technologies, and in particular to a method, device and computing device for using a microphone to track a speaker. Background technique [0002] At present, speech recognition requires the front-end microphone to collect data for processing. However, in a complex environment, the effective voice data collected by the microphone is incorrectly collected and distorted. Different angles and distances cause a significant gap in the effective voice data collection effect of the speaker. [0003] In the prior art, the position of the traditional microphone is fixed, and the debugging and development of various angles and distances need to be considered comprehensively. The resulting development cost and time cost are very high, and in order to be compatible with voice collection and recognition from various angles, it may It will sacrifice more performance or computing power, or recognit...

AI Technical Summary

Problems solved by technology

[0002] At present, speech recognition requires front-end microphones to collect data for processing, but in a complex environment, the effective speech data collected by the microphone has miscollection and distortion, different angles and different distances, and the effective speech data collection effect of the speaker is significantly different.

[0003] In the existing technology, the position of the traditional microphone is fixed, and the debugging and development of various angles and distances need to be considered comprehensively. It will sacrifice more performance or computing power, or recognition effect, and the final effect is not the most ideal state in the research and development process

Images