Abstract
Voice-based interactions frequently experience suboptimal audio quality because default audio input selection relies on static rules or basic heuristics rather than contextual variables and user history. A technique is described wherein a large language model is utilized to dynamically select an optimal audio input source from multiple connected devices. Contextual data, including device characteristics, historical manual switching behavior, application types, and audio quality feedback, is provided to the large language model to determine the appropriate audio input. By continuously adapting based on user interactions and overrides, optimal audio quality is consistently maintained, and the necessity for manual audio source configuration is minimized.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Zimmermann, Nicolas, "Intelligent Audio Input Selection Using Large Language Models", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10895