US 2026/0268911 A1 · METHOD AND SYSTEM FOR ADAPTIVELY ACQUIRING IMAGE DATA FOR PERSONALIZED SPEECH DIALOGUES

Abstract

A method and a system for adaptively acquiring image data for personalized speech dialogue. In the method, the system receives speech data generated by a user in real time from a user device and extracts speech features from the speech data. A natural language processing model is used to recognize semantic meaning in the speech features. The system then determines that an image is required to be provided. A function of photographing images in the user device is activated for acquiring image data. Both the speech features and image features are referred to for generating corresponding speech dialogue. After repeatedly the above steps such as receiving speech data generated by the user from the user device, extracting speech features, receiving image data generated by the user device, and extracting image features, the speech dialogues based on the speech features and the image features can be continuously generated.

Claims20

Dependent claims

A method for adaptively acquiring image data for personalized speech dialogues, operated in a cloud server, comprising:

receiving a speech data generated by a user in real time via a user device, extracting current speech features of the user, and recognizing semantics of the speech features by a natural language processing model;
activating a function for capturing images in the user device after determining that the images are required to be captured so as to obtain an image data, and generating speech dialogues data based on the speech features and the image features;
and
continuously generating the speech dialogues based on the speech features and the image features after repeatedly receiving the speech data generated by the user via the user device, extracting the speech features, receiving the image data generated by the user device and extracting the image features.

1 more element

A system, which is used to operate a method for adaptively acquiring image data for personalized speech dialogues, comprising:

a cloud server, connecting with a user device, providing a personalized speech dialogue service to a user via a dialogue interface and performing the method for adaptively acquiring image data for personalized speech dialogues by one or more processors, wherein the method comprises:
receiving a speech data generated by the user in real time via the user device, extracting current speech features of the user, and recognizing semantics of the speech features by a natural language processing model;
activating a function for capturing images in the user device after determining that the images are required to be captured so as to obtain an image data, and generating speech dialogues data based on the speech features and the image features;
and
continuously generating the speech dialogues based on the speech features and the image features after repeatedly receiving the speech data generated by the user via the user device, extracting the speech features, receiving the image data generated by the user device and extracting the image features.

2 more elements

Dependencies are inferred from the claim text.

Method and system for adaptively acquiring image data for personalized speech dialogues

Life of this patent

Method and system for adaptively acquiring image data for personalized speech dialogues

Life of this patent

Abstract

Claims20