GOOGLE LLC (20240242719). HOTWORD SUPPRESSION simplified abstract
Contents
HOTWORD SUPPRESSION
Organization Name
Inventor(s)
Alexander H. Gruenstein of Mountain View CA (US)
Taral Pradeep Joglekar of Sunnyvale CA (US)
Vijayaditya Peddinti of San Jose CA (US)
Michiel A.U. Bacchiani of Summit NJ (US)
HOTWORD SUPPRESSION - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240242719 titled 'HOTWORD SUPPRESSION
The abstract describes a method where a first computing device adds an audio watermark to speech data containing a hotword to get the attention of a second computing device. The second device receives the watermarked speech data and decides to stop processing it.
- The method involves adding an audio watermark to speech data with a hotword to alert another device.
- The first computing device plays back the speech data with the watermark.
- The second computing device receives the watermarked speech data and stops processing it.
Potential Applications: This technology could be used in voice-activated devices to improve user experience and privacy by signaling when certain commands are being used.
Problems Solved: This technology helps in efficiently managing speech data processing and enhances the security of voice-activated systems.
Benefits: Enhanced user experience, improved privacy, and better security in voice-activated devices.
Commercial Applications: Title: Voice-Activated Device Security Enhancement This technology could be valuable for companies developing voice-activated devices, smart speakers, and virtual assistants to enhance security and user experience.
Questions about Voice-Activated Device Security Enhancement: 1. How does the audio watermark help in improving the security of voice-activated devices? The audio watermark alerts the second computing device to stop processing the speech data, enhancing security by preventing unauthorized access to sensitive information.
2. What are the potential implications of this technology for the future of voice-activated systems? This technology could lead to more secure and efficient voice-activated devices, improving user trust and adoption in the market.
Original Abstract Submitted
a method includes adding, by a first computing device, a first audio watermark to first speech data corresponding to playback of a first utterance including a hotword used to invoke an attention of a second computing device. the method includes outputting, by the first computing device, the playback of the first utterance corresponding to the watermarked first speech data. the second computing device is configured to receive the watermarked first speech data and determine to cease processing of the watermarked first speech data.
