Google patents voice assistant that asks before acting on commands
This patent describes a method where a device listens for a specific wake word and a following command. Instead of immediately executing the command, the device sends a notification asking the user to confirm the action. If the user does not confirm, or if the device detects a negative response, it cancels the command.
In the filing’s own words
A method for a soft acceptance of a hotword receives audio data characterizing a soft hotword event detected by a hotword detector in streaming audio captured by a user device. The method also processes the audio data to determine that the audio data corresponds to a query specifying an action to perform on the user device. Without triggering performance of the action on the user device or the other device, the method provides a notification for output from the user device where the notification prompts a user associated with the user device to provide an affirmative input indication in order to trigger performance of the action on the user device or the other device and, when the user fails to provide the affirmative input indication, instructs the user device or the other device to not perform the action specified by the query.
What exactly they claim (claim 1)
1. A computer-implemented method executed by data processing hardware that causes the data processing hardware to perform operations comprising: receiving audio data captured by a user device, the audio data comprising a first audio segment and a second audio segment, the first audio segment comprising a hotword configured to initiate processing of one or more terms following the hotword, the second audio segment comprising an action to perform on the user device or another device; determining that the first audio segment characterizes a soft hotword event; based on determining that the first audio segment characterizes the soft hotword event, determining that the second audio segment corresponds to a query specifying the action to perform on the user device or the other device; and based on determining that the first audio segment characterizes the soft hotword event and determining that the second audio segment corresponds to the query specifying the action to perform on the user device or the other device: providing a notification for output from the user device, the notification prompting a user associated with the user device to provide an affirmative input indication in order to trigger performance on the user device or the other device; activating a warm word model configured to detect, without performing speech recognition, a presence of both of: one or more spoken terms characterizing the affirmative input indication; and one or more other spoken terms characterizing a negative input indication, the negative input indication indicating that the user does not want the user device or the other device to perform the action specified by the query; receiving a speech input comprising the one or more other spoken terms characterizing the negative input indication; detecting, by the warm word model in the speech input, without performing speech recognition, the one or more other spoken terms characterizing the negative input indication; determining that the user failed to provide the affirmative input indication based on detecting the one or more other spoken terms characterizing the negative input indication; and based on determining that the user failed to provide the affirmative input indication, instructing the user device or the other device to not perform the action specified by the query.
Facts
- Company
- Google (Google LLC)
- Official title
- Hotwording by degree
- Patent number
- US12725609B2
- Application number
- 18735706
- Filed
- Jun 6, 2024
- Granted
- Sep 1, 2026
- Inventors
- 3
- Claims
- 12
Classified under
G10L 15/08 Physics & computingG10L 15/22 Physics & computingG10L 2015/088 Physics & computingMore from Google in Audio & voice
- Google patents voice assistant that cuts off its own answer to hear a new question Sep 15, 2026
- Google patents head-worn device that cleans up voice using bone vibration Sep 17, 2026
- Google patents ring design to smooth out microphone sound quality Sep 1, 2026
- Google patents earbud part that combines sound and vibration Sep 8, 2026
- Google patents AI that identifies speakers in long audio clips Sep 22, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.