Abstract
When a user moves in physical space while engaging in conversational hands-free interaction with a voice-based virtual assistant on a device, the interaction is interrupted if the user steps too far from the device. Currently, devices that support such interaction do not provide dynamic audio switching or mechanisms to switch an ongoing conversational interaction to a different device. This disclosure describes techniques for seamless dynamic switching of audio input and output from one device to another based on presence detection using data from device sensors. The appropriate devices for the audio input and the output, as well as the device that acts as the host of the virtual assistant can be determined by following any suitable arbitration procedure, guided by prespecified or inferred user preferences for virtual assistant interaction. Automated dynamic remapping of audio input and out and/or virtual assistant host device can enhance the user experience (UX) by enabling users to engage in seamless and fluid conversation interactions with a virtual assistant while moving around.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Shin, D, "Contextual Remapping of Audio Input and Output for Fluid Virtual Assistant Interaction", Technical Disclosure Commons, (August 03, 2023)
https://www.tdcommons.org/dpubs_series/6116