Personalized gesture recognition for user interaction with assistant systems
In one embodiment, a method includes receiving a user request from a first user from a client system associated with a first user, wherein the user request comprise a gesture-input from the first user and a speech-input from the first user, determining an intent corresponding to the user request based on the gesture-input by a personalized gesture-classification model associated with the first user, executing one or more tasks based on the determined intent and the speech-input, and sending instructions for presenting execution results of the one or more tasks to the client system responsive the user request.
1. A system comprising:
one or more processors;
a microphone, operably coupled to at least one of the one or more processors, configured to generate a speech-input based on an audible signal from a user;
a camera, operably coupled to at least one of the one or more processors, configured to generate a gesture-input based on an action performed by the user; and
one or more non-transitory non-volatile memories, operably coupled to at least one of the one or more processors, comprising instructions that, when executed by at least one of the one or more processors, cause the system to:
communicate a user input comprising at least one of speech information or gesture information to an external assistant system to cause the external assistant system to determine an output user intent from the user input, wherein the speech information is based at least in part upon the speech-input, the gesture information is based at least in part upon the gesture-input, and the output user intent is based at least in part upon the speech information or gesture information;
in response to communicating the user input, receive, from the external assistant system, information associated with the output user intent; and
execute a task based at least in part upon the received information, wherein the task is based at least in part upon the speech-input and the gesture-input.
2. The system of claim 1 , wherein the speech-input comprises audio information.
3. The system of claim 1 , wherein the speech information is based on converting the speech-input using an automatic speech recognition (ASR) module.
4. The system of claim 1 , wherein the action is performed using the user's eyes.
5. The system of claim 1 , wherein the gesture-input comprises image information, video information, or motion information.
6. The system of claim 1 , wherein the gesture information comprises a feature representation for a gesture, associated with the action performed by the user, wherein the feature representation for the gesture is generated based at least in part upon the gesture-input.
7. The system of claim 1 , wherein the user input comprises the speech information and the gesture information.
8. The system of claim 1 , wherein the output user intent is based on the speech information and the gesture information.
9. The system of claim 1 , wherein the task is based at least in part upon a user profile associated with the user.
10. The system of claim 1 , further comprising a virtual reality (VR) headset comprising a display, wherein execution of the task causes a content item to be displayed on the display of the VR headset.
11. The system of claim 1 , wherein the gesture information is based at least in part upon a machine-learning model.
12. The system of claim 11 , wherein the machine-learning model is personalized for the user.
13. A method comprising:
generating a speech-input based on an audible signal from a user captured by a microphone;
generating a gesture-input based on an action performed by the user captured by a camera;
communicating a user input comprising at least one of speech information or gesture information to an external assistant system to cause the external assistant system to determine an output user intent from the user input, wherein the speech information is based at least in part upon the speech-input, the gesture information is based at least in part upon the gesture-input, and the output user intent is based at least in part upon the speech information or gesture information;
in response to communicating the user input, receiving, from the external assistant system, information associated with the output user intent; and
executing a task based at least in part upon the received information, wherein the task is based at least in part upon the speech-input and the gesture-input.
14. The method of claim 13 , wherein the speech information is based on converting the speech-input using an automatic speech recognition (ASR) module.
15. The method of claim 13 , wherein the action is performed using the user's eyes.
16. The method of claim 13 , wherein the gesture information is based at least in part upon a machine-learning model.
17. A non-transitory non-volatile computer-readable medium operably coupled to one or more processors, comprising instructions that, when executed by at least one of the one or more processors, cause the at least one of the one or more processors to:
generate a speech-input based on an audible signal from a user captured by a microphone;
generate a gesture-input based on an action performed by the user captured by a camera;
communicate a user input comprising at least one of speech information or gesture information to an external assistant system to cause the external assistant system to determine an output user intent from the user input, wherein the speech information is based at least in part upon the speech-input, the gesture information is based at least in part upon the gesture-input, and the output user intent is based at least in part upon the speech information or gesture information;
in response to communicating the user input, receive, from the external assistant system, information associated with the output user intent; and
execute a task based at least in part upon the received information, wherein the task is based at least in part upon the speech-input and the gesture-input.
18. The non-transitory non-volatile computer-readable medium of claim 17 , wherein the speech information is based on converting the speech-input using an automatic speech recognition (ASR) module.
19. The non-transitory non-volatile computer-readable medium of claim 17 , wherein the action is performed using the user's eyes.
20. The non-transitory non-volatile computer-readable medium of claim 17 , wherein the gesture information is based at least in part upon a machine-learning model.