A wireless earphone and noise reduction method thereof

Abstract

Embodiments of the present disclosure provide a wireless earphone and a noise reduction method thereof, where the wireless earphone includes a microphone, an interface, and a memory. The microphone is configured to collect audio signals; the interface is configured to receive the collected audio signals; and multiple learning networks are respectively stored on the memory corresponding to multiple noise scene categories. The wireless earphone also includes a processor configured to: classify the received audio signal to determine the category of the noise scene it is in; based on the determined category of the noise scene, call the corresponding learning network to The signal undergoes noise reduction processing. The wireless headset can achieve the noise reduction effect that a large-scale learning network can achieve through a small learning network, and reduce computing resources and increase computing speed by reducing the computational complexity of a deep neural network. In addition, the above technical solution avoids the requirements on the cavity and size of the earphone, and reduces the production cost of the earphone.

Description

technical field [0001] The present disclosure relates to the field of communication equipment, in particular to a wireless earphone and a noise reduction method thereof. Background technique [0002] With the progress of society and the improvement of people's living standards, earphones have become an indispensable daily necessities in people's lives. People often answer and make calls on the street, bus, and subway. The noise in this environment will be picked up by the microphone of the earphone, mixed with the content of our speech, and sent upstream to the earpiece of the other party's phone, thus It affects the voice quality of the other party's answering the call, so the noise reduction technology of the headset is very necessary. [0003] At present, the performance and complexity of the noise reduction algorithm of the wireless Bluetooth headset directly affect the power consumption, cost and user experience of the headset. Therefore, in order to achieve a better n...

AI Technical Summary

Problems solved by technology

This solution has a better noise reduction effect on ambient noise and directional interference sound sources, but the disadvantage is that it relies on the phase difference between the two microphones to reduce noise, which has a negative effect on the cavity design and the consistency between the two microphones. Higher requirements increase the design complexity and product cost, and a certain distance is required between the two microphones to ensure the noise reduction effect. However, the smaller the size of the current earphones, the more conducive to market promotion, so the dual-mic earphone solution gradually meets Can't understand the distance requirements of the dual microphone noise reduction algorithm

Images