11 results about "Secure voice" patented technology

This invention discloses a secure voice-based intelligent customer service system and interaction method based on a multi-agent architecture. The system includes an ASR (Automatic Speech Recognition) module, multi-layered security barriers, a main retrieval agent, a specialized retrieval agent cluster, a parallel retrieval and result aggregation module, an LLM (Limited Language Modeling) synthesis engine, and a TTS (Text-to-Speech) module. Through multi-agent system architecture and methodological design, this invention achieves modular processing, parallel retrieval tasks, and closed-loop security control. By enabling independent calls between input, output, agents, and inference stages via multi-layered security barriers, the system effectively mitigates risks such as prompt word injection, indirect injection attacks, harmful content generation, and information leakage. Sensitive information identification, access permission verification, and content compliance review are embedded in each stage of input, retrieval, information synthesis, and output to prevent sensitive data leakage and meet the application requirements of high-security business scenarios (such as finance, telecommunications, and government services).

This invention provides an end-to-end voice encryption method, apparatus, and Bluetooth headset suitable for multi-hop lossy channels. The method includes front-end noise reduction, framing, and segmentation of the acquired voice at the transmitting end. Based on the session mode, a group public key or a point-to-point private key is automatically selected as the master key from a pre-configured key set. Then, multiple time-domain segments within each frame are scrambled, subjected to segment-by-segment Fast Fourier Transform, and frequency-domain encryption based on subkeys, according to the master key. The encrypted voice is then generated through inverse transformation and overlapping weighted smoothing concatenation, and transmitted through a multi-hop voice communication channel containing multi-level lossy encoding and decoding. At the receiving end, the reverse process is performed: framing and segmentation, transformation, inverse frequency-domain encryption, and inverse time-domain scrambling. Combined with U-shaped neural network noise reduction at both the front-end and back-end, the voice is restored to intelligible speech. This method enables flexible key management and high-quality secure voice transmission in scenarios where group calls and point-to-point private calls coexist and undergo multiple lossy compression operations.

The invention discloses a transformer substation operation safety voice prompt method and system based on multi-mode perception, and relates to the technical field of voice prompt, and the method comprises the following steps: collecting the position and operation action information of an operator in real time, and judging the operation danger level; environment electromagnetic interference signals are collected, and voice prompt triggering conditions are generated in combination with the operation danger level; generating voice signal reconstruction aiming at the current environment interference based on the voice prompt triggering condition; based on the reconstructed voice signal and the danger level, the most suitable loudspeaker is selected to carry out voice broadcasting, so that a high-danger event prompt is preferentially transmitted to a target operator; after voice broadcasting, response and action changes of an operator are monitored, and voice signal characteristics and a broadcasting sequence are dynamically adjusted. According to the invention, by sensing the operation action and the environment interference in real time, dynamically generating the adaptive voice prompt and adjusting the broadcast sequence and the signal characteristics in a closed loop, the problems of prompt delay and report omission under strong interference are solved, and the operation safety is improved.

Systems and methods of the present disclosure enable secure voice control of a renal care device. A controller in communication with the renal care device is configured to receive a voice signal, input the voice signal into a voice recognition machine learning model to output a voice authentication determination and a device command based on trained parameters of the voice recognition model. The trained parameters of the voice recognition model are configured to determine a measure of a match between the voice signal and at least one training vocal pattern of a training vocal pattern dataset. Based on the voice authentication determination and the device command, the controller is configured to instruct the renal care device with the device command based at least in part on the measure of the match of the voice signal exceeding a predetermined threshold.

1. The name of the design product: graphical user interface of AI companion check-in gift for electronic device. 2. The use of the design product: the design product is used for displaying information, running programs and communication. 3. The design points of the design product: the graphical user interface content in the screen. 4. The picture or photo that best indicates the design points: reference figure 2 for use state. 5. The use of the graphical user interface: for showing the guidance interface of AI companion to guide users to check in and send gifts. 6. The human-computer interaction mode of the graphical user interface: the main view is the main interface of AI companion, and the bottom operation bar includes four buttons of call record, IM chat, good night and security voice from left to right; the use state can be referred to the use state reference figure 1. When the first "gift" button on the left in the upper right corner is clicked, the graphical user interface jumps from the main view to the interface change state figure 1, showing the AI companion check-in pop-up interface; the use state can be referred to the use state reference figure 2, and the gift list is displayed on the interface. Click the gift picture on the interface change state figure 1, the graphical user interface jumps from the interface change state figure 1 to the interface change state figure 2, showing the user selected gift picture; the use state can be referred to the use state reference figure 3, and the marks ①②③… are displayed on the user selected cute pet gift. Click the "choose it" button at the bottom of the interface change state figure 2, the graphical user interface jumps from the interface change state figure 2 to the interface change state figure 3, showing the check-in interface; the use state can be referred to the use state reference figure 4, and the fragments can be obtained. Click the "check in" button at the bottom of the interface change state figure 3, the graphical user interface jumps from the interface change state figure 3 to the interface change state figure 4, showing the obtained fragments; the use state can be referred to the use state reference figure 5. Click the "receive" button at the bottom of the interface change state figure 4, the graphical user interface jumps from the interface change state figure 4 to the interface change state figure 5, showing the collected fragments; the use state can be referred to the use state reference figure 6. If all the fragments have been collected, the "get reward" button on the interface change state figure 5 can be clicked to get the reward.

The invention discloses a multilingual medical voice commerce system enabling secure, voice-activated transactions for healthcare products, services, and prescriptions. Utilizing AI-driven speech recognition optimized for medical terminology, the system integrates natural language processing and context-aware recommendation engines. Secure features include HIPAA-compliant authentication, patient consent management, and end-to-end encryption. The platform interfaces with electronic health records, pharmacy networks, insurance verification systems, and wearable health monitors for real-time medical data access and transaction execution. Designed for hospital procurement, surgical environments, telehealth, and home healthcare, the system incorporates universal accessibility, including adaptations for visual, motor, cognitive, and speech impairments. Multilingual capabilities ensure cultural and linguistic inclusivity. AI analytics track usage patterns, optimize recommendations, detect health trends, and support population health initiatives. Operable across smart speakers, medical devices, and connected platforms, the system delivers a unified, monetized, secure, and inclusive medical voice commerce ecosystem for patients, providers, and healthcare organizations worldwide.

The invention provides an end-to-end voice encryption method and device suitable for a multi-hop lossy channel and a Bluetooth headset, and the method comprises the steps: carrying out the front-end noise reduction, framing and segmentation of collected voice at a transmitting end, automatically selecting a group public key or a point-to-point private key from a pre-configured key set as a main key according to a session mode, and carrying out the end-to-end voice encryption of the multi-hop lossy channel. And performing time sequence scrambling, segment-by-segment fast Fourier transform and sub-key-based frequency domain encryption on a plurality of time domain sub-segments in each frame according to the master key, generating encrypted voice through inverse transformation and overlapped weighted smooth splicing, and transmitting the encrypted voice through a multi-hop voice communication channel containing multi-stage lossy coding and decoding. Framing segmentation, transformation, inverse frequency domain encryption and inverse time domain scrambling which are opposite to the above process are executed at a receiving end, and noise reduction is combined with the U-shaped neural networks at the front end and the rear end to recover the voice to be understandable. According to the invention, flexible key management and high-quality secure voice transmission can be realized in a scene in which a group call and a point-to-point private call coexist and are subjected to lossy compression for multiple times.

The utility model relates to the technical field of audio playing, and discloses a factory safety voice playing device, which comprises a control panel, a voice storage module, a voice recognition module, a power supply module and at least one audio output module. A plurality of audio trigger keys are arranged on the control panel, and the audio trigger keys are used for responding to pressing operation of a user and outputting corresponding audio selection signals; the voice storage module is used for storing a plurality of segments of safe audio; the voice recognition module is respectively connected with the plurality of audio trigger keys and the voice storage module, and the voice recognition module is used for recognizing a corresponding safe audio according to the audio selection signal; and the audio output module is connected with the voice recognition module and is used for amplifying the corresponding safe audio so as to play the corresponding safe audio. The played safe audio can be conveniently switched, the sensitivity is improved, the response time is shortened, the self-recognition of the safe audio is realized, the misoperation is reduced, and the production safety is improved.

An intelligent, humane animal deterrent system utilizing AI and multivision sensor fusion for real-time species recognition and behavioral threat assessment. The system integrates thermal and RGB cameras, motion and acoustic sensors, RFID scanning, and secure voice command. Non-lethal directional deterrents—ultrasonic sound, LED arrays, and vibration—are mounted on precision swivel-tilt mechanisms for dynamic targeting. Modular deployment supports handheld, wearable, fixed, and drone-mounted configurations via universal, tool-free attachments. AI processing is hybrid, combining edge computing and cloud-based neural networks. A hybrid communication module enables mesh networking, satellite links, and emergency beacons with automatic fallback. Power is managed by dual batteries and optional solar charging. Smart displays and AR eyewear offer real-time ID and safety guidance. Voice commands support hands-free, multi-language use. The system improves personal safety, protects property, and promotes humane wildlife interaction and conservation.

Embodiments are provided for protecting data access to machine learning training data at multiple distributed computing devices. Electronic content comprising raw data corresponding to a preferred data security level is partitioned into multiple micro segments. The multiple micro segments are distributively distributed to multiple computing devices that apply transcription labels to the multiple micro segments. The labeled micro segments are reconstructed into training data that is subsequently used to train a machine learning model while facilitating an increase in data security of the raw data included from the reconstructed micro segments.

1. Name of this design product: Intelligent Voice-activated Safety Alarm Device for Power Facilities. 2. Purpose of this design: For use in broadcasting safety voice warnings for power facilities. 3. The key design features of this product are the combination of shape and pattern. 4. The image or photograph that best illustrates the design's key features: the front view.