Samsung patents voice assistant fix for when you say the wrong thing
This patent describes a method for electronic devices to handle voice command errors. When a user speaks a command, the device picks the most likely match, but if the user immediately speaks again to correct it, the system identifies the new phrase as a fix. It then ignores the first command and executes the second, more accurate command instead.
In the filing’s own words
According to an embodiment, an electronic device comprises a microphone and at least one processor operatively connected with the microphone, wherein the processor is configured to obtain one or more commands corresponding to a first utterance obtained through the microphone, based on the first utterance, execute an operation corresponding to a first command having a highest confidence score for the first utterance among the one or more commands, identify that a second utterance obtained through the microphone is an utterance for correcting the operation corresponding to the first command, based on the second utterance, determine a second command having another highest confidence score for the second utterance among the one or more commands except for the first command, based on identifying that the second utterance is the utterance for correcting the operation corresponding to the first command, and execute another operation corresponding to the second command that is determined.
What exactly they claim (claim 1)
1. An electronic device comprising: a microphone; at least one processor operatively connected with the microphone; and memory storing instructions that, when executed by the at least one processor individually or collectively, cause the electronic device to: obtain one or more commands corresponding to a first utterance obtained through the microphone, based on the first utterance, execute first operation corresponding to a first command having a highest confidence score for the first utterance among the one or more commands, wherein the one or more commands and corresponding confidence scores for the first utterance are stored in the memory, identify that a second utterance obtained through the microphone is an utterance for correcting the first operation corresponding to the first command, based on the second utterance, determine a second command having a highest confidence score for the second utterance among the one or more commands excluding the first command, based on identifying that the second utterance is the utterance for correcting the first operation corresponding to the first command, wherein determining the second command having the highest confidence score for the second utterance comprises: retrieving from the memory the one or more commands and the corresponding confidence scores for the one or more commands, and comparing the second utterance to the one or more commands excluding the first command to obtain the highest confidence score for the second utterance among the one or more commands retrieved from the memory, such that the second command is identified, and obtain a first label and a second label from the second utterance, using a pre-trained utterance identification model, identify whether the second utterance is the utterance for correcting the first operation corresponding to the first command, based on the first label, identify a correction phrase included in the second utterance, based on the second label obtain a confidence score for the one or more commands excluding the first command and the correction phrase, perform the second operation corresponding to the second command having the highest confidence score for the correction phrase among the one or more commands excluding the first command, and execute a second operation corresponding to the second command.
Facts
- Company
- Samsung (SAMSUNG ELECTRONICS CO., LTD.)
- Official title
- Method for analyzing user utterance and electronic device supporting the same
- Patent number
- US12738270B2
- Application number
- 18236831
- Filed
- Aug 22, 2023
- Granted
- Sep 15, 2026
- Inventors
- 1
- Claims
- 18
Classified under
G10L 15/22 Physics & computingG10L 15/1815 Physics & computingG10L 2015/223 Physics & computingG10L 15/30 Physics & computingMore from Samsung in Audio & voice
- Samsung patents audio recovery that rebuilds sound from visual data Sep 8, 2026
- Samsung patents earbuds that switch to singing mode when you start singing Sep 8, 2026
- Samsung patents sound masking that adapts to your surroundings Oct 13, 2026
- Samsung patents method to fix audio distortion for better echo cancellation Sep 1, 2026
- Samsung patents headphones that fix noisy music using a server Sep 15, 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.