Abstract
When one is in transit, different kinds of non-stationary noises may arise. Moreover, in different countries non-stationary noise intensity levels may vary significantly from location to location. For example, in a market in India the intensity of noise will be much higher than that found in a market in the United States. Currently, however, artificial intelligence (AI) -driven headsets are trained on a generic noise-related dataset that is used to filter out noise. To address these types of challenges, techniques are presented herein that support an AI-based, state of the art, intelligent, and interactive algorithm that will detect current headset location based on a geolocation tag and invoke the appropriate specialized purposely-built pre-trained flows to cancel or suppress noise.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Mukherjee, Anupam and Jain, Vibhor, "GEOLOCATION DRIVEN REINFORCMENT LEARNING-POWERED IN TRANSIT HEADSET NOISE CANCELLATION MECHANISM", Technical Disclosure Commons, (March 05, 2021)
https://www.tdcommons.org/dpubs_series/4133