A method and system for storing electronic documents based on a distributed environment includes the following steps of: a storage management server receives a document data stream sent by a client; the storage management server parses the document type of the electronic document; the storage management server Match the corresponding first storage server based on the document type of the electronic document and the user information; the storage management server sends the corresponding electronic document to the corresponding first storage server; the first storage server is based on the first storage server The digital watermark loading rule of the server loads the digital watermark on the electronic document; the first storage server stores the electronic document loaded with the digital watermark. The embodiment of the present invention facilitates the distributed storage server to store different types of electronic documents in combination with electronic document types, and ensures that subsequent users can manage and refer to the stored electronic documents.

A method and apparatus for digital watermarking, recording multimedia contents, as well as detection, by using data recorded in a base, of potentially infringing multimedia contents. The method is based on a watermarking done in conjunction with the recording of specific data in a data base, this data being thereafter used to detect a watermark, if any, in an infringing content in using or not using the original content.

The present invention proposes a rapid, reliable data extraction method for extracting data from a database of data (3) that is watermarked according to a reversible watermarking method. The method according to the invention consists of obtaining, from a first extraction request received from a client terminal (1), a second extraction request (R2) for extracting watermarked data and a third request (R3) applied to the extracted data after it has been de-watermarked. In the presence in the first extraction request of initial selection conditions relating to watermarked properties, these are included in the second request after having been transformed. In the presence in the first extraction request (R1), of aggregation functions relating to a watermarked extraction property, these are included without modification in the third request (R3) so as to supply a result to the client terminal (1) in response to the first extraction request (R1).

Embodiments of the disclosure relate to learning new watermark algorithms for artificial intelligence (AI) models for a data processing (DP) accelerator. In one embodiment, a system trains a watermark algorithm based on a predetermined set of criteria, where the watermark algorithm is trained to generate variations of the watermark algorithm. The system configures the watermark unit at runtime with a variation of the watermark algorithm for the watermark algorithm to be used by the DP accelerator.

A device for documenting and encrypting a face to face personal meeting that includes a casing, a connecting unit that can communicate with external data storage device, a camera or microphone, an encryption subsystem, a decryption subsystem, and a personal key reader that is capable to identify personal keys of participants of the meeting. The device can encrypt the video that can be decrypted by using the personal keys of the participants. The device can save the encrypted video only in external storage device that communicates with the connecting unit while the camera shoots the video, and it is saved with information of the participants' identified personal keys. The decryption subsystem can decrypt the encrypted video when external storage device in which the encrypted video is saved communicates with the connecting unit and when the reader identifies the identified personal keys of the participants.

A plurality of byte ranges forms a sample for content output from a player device, and includes at least one double-encrypted byte range. The plurality of byte ranges is stored in a secured memory, and the at least one double-encrypted byte range is partially decrypted to generate at least one decrypted singe-encrypted byte range. The plurality of byte ranges is stored in an unsecured memory using the at least one decrypted single-encrypted byte range in place of the at least one double-encrypted byte range.

Data-analysis-based processing is provided to validate audio content delivery via embedded inaudible sound signals. The processing obtains, by a validation system, an indication of detection of an inaudible sound signal embedded in an audio output of provided content initiated by a user on a user-device. The inaudible sound signal is inaudible to the user, and the audio output is from the user-device. Based on obtaining the indication of detection of the inaudible sound signal embedded in the audio output, the validation system validates that the audio output of the provided content is ON with respect to the user-device.

Systems, methods, and computer-readable media for controlling the execution of a deployed software program to a customer system are disclosed herein. A vendor may deploy a software program to the customer system. The software program may comprise an access key, the access key comprising a digital signature and access parameters. The digital signature may utilize a public key private key pair. The customer may run the software program by validating the access key on the customer system. Validating the access key may comprise verifying the digital signature and verifying the access parameters. Once the access key has been validated, the customer may execute the software program on the customer system.

The present disclosure provides a voice authentication system. The voice authentication system according to an embodiment of the present disclosure includes a voice collection unit configured to collect voice information obtained by digitizing a speaker's voice, a learning model server configured to generate a voice image based on the collected voice information of the speaker, causes a deep neural network (DNN) model to learn the voice image, and extract a feature vector for the voice image, a watermark server configured to generate a watermark based on the feature vector and embed the watermark and individual information into the voice image or voice conversion data, and an authentication server configured to generate a private key based on the feature vector and determine whether to extract the watermark and the individual information based on an authentication result.

Content screening operations, which can include watermark extraction and the application of content usage enforcement actions, may be organized such that some or all of the operations can be conducted by different devices that are connected via connectors such as HDMI (High-Definition Multimedia Interface (HDMI), analog composite video, DVI (Digital Visual Interface), SDI (Serial Digital Interface), DisplayPort, or networked via Ethernet or wireless. Authentication and encryption methods are disclosed that can be used to establish the trust and secure communication between devices that conduct collaborative content screening. Delegation architecture may be based on ascertained screening capabilities of the sink device wherein the source device verifies that the sink device is capable and trusted to perform partial or whole screening operations delegated by the source. Alternatively, delegation architecture may be based on ascertained screening capabilities of the source device wherein the sink device verifies that the source device is capable and trusted to provide correct content credential and content use policy (“content credential”) that is needed for the sink device to determine whether content screening should be performed, and if yes, what content screening operations should be performed.