749 results about "Voice assistant" patented technology

The invention relates to a data access method and system of an intelligent Internet of Things gateway. The method comprises the following steps: performing feature matching and protocol identification on a plurality of sensors and terminal equipment connected to the intelligent Internet of Things gateway by using an equipment feature library to obtain an equipment protocol type; analyzing the original data to obtain standardized data; performing encryption processing to obtain an encrypted security data packet and an encrypted session key; performing multi-factor transmission channel selection to obtain an optimal transmission channel combination; and based on the optimal transmission channel combination, sending the encrypted security data packet to the Internet of Things platform, verifying the received data confirmation information based on a consensus algorithm, and outputting a data access state report through a data visual interface and an AI voice assistant. The processing strategy can be dynamically adjusted according to the equipment characteristics, the data characteristics and the network environment, and the flexibility and adaptability of the system are improved.

The invention provides a Multi-Agent-based voice interaction method, which comprises the following steps of S1, waking up a voice assistant, collecting voice data input by a user, converting the collected voice data into text input through an STT voice-to-text module, and storing the text input into a historical dialogue database; through a weighted intention classification algorithm using an intention classification large model and a preset vector library, accurate distribution of user tasks is significantly improved. Meanwhile, the intelligent home and equipment control are set as independent agents, the burden of an intention classification large model is relieved, and the intelligence and accuracy of the system are improved. The design ensures that a system link is simple and efficient, provides highly personalized services, adapts to various devices and application scenes, and improves user experience.

The invention discloses an AR glasses control system and method, and belongs to the field of artificial intelligence and AR glasses application, and the method comprises the steps: capturing a user face image, combining a voice assistant, collecting user voiceprint data, and carrying out the matching through a recurrent neural network; constructing a three-dimensional environment model, positioning and marking obstacles and interactive objects, and identifying objects in the environment; the method comprises the following steps of: tracking a gesture, acquiring a key point coordinate by adopting an OpenPose algorithm, identifying a predefined gesture by utilizing a support vector machine, and mapping the gesture into a specific operation instruction; analyzing the user intention based on natural language processing, and calling a corresponding function module according to an analysis result; acquiring a gazing direction of a user by using an eye movement tracking sensor, calculating a screen focus position through an algorithm, and dynamically highlighting options or displaying information according to a gazing point position; through multi-modal data fusion, gestures, voices and fixation point input are analyzed in a unified manner. By accurately understanding the intention of the user, the interaction precision of the AR glasses is improved, and the user experience is optimized.

The invention discloses an intelligent infusion monitoring management system, and relates to the technical field of medical equipment, and the system comprises the steps: collecting patient data, synchronizing the patient data to a central monitoring system through a block chain technology, and generating a personalized infusion scheme; based on a personalized infusion scheme, a medical worker uses a mobile device to carry out identity verification on a patient and obtain medicine information, and a voice assistant function is integrated; the monitoring module is responsible for monitoring vital signs of a patient and the temperature of a ward by using wearable equipment and a temperature sensor, and adjusting a personalized infusion scheme and liquid temperature according to the vital signs and the temperature of the ward; the risk value prediction module is responsible for predicting risk values of blockage and leakage according to the adjusted personalized infusion scheme and the liquid temperature and setting a graded alarm mechanism; patients and family members receive the infusion progress through the intelligent infusion monitor. According to the invention, the block chain technology and the advanced artificial intelligence algorithm are integrated, so that the health data and medicine information of the patient can be efficiently managed and analyzed.

The invention discloses an end-to-end deep counterfeit voice detection system and method, and belongs to the technical field of voice processing. The detection system is divided into a front-end embedded extraction network and a rear-end classification network. And the front-end embedded extraction network adopts a front-end pre-training model to extract global semantic features of different scales of the voice signals, and adopts a feature fusion module to perform deep fusion on the extracted global semantic features of different scales, so that the expression ability of feature embedding is further enhanced. And the rear-end classification network adopts a single-layer long-short-term memory classification network to classify the fusion features, performs joint optimization by using cross entropy loss and center loss, and finally judges whether the voice is real or forged. The method is applied to scenes of voice security, identity authentication, voice assistants and the like, a more accurate and robust deep counterfeit voice detection scheme with good generalization ability can be provided, and the method has important significance for improving the security and credibility of a voice recognition system.

Dialogue state tracking for voice assistants involves correctly tracking intent and entities of a task that a user is performing. A dialogue state, having a tracked intent and one or more tracked entities, can then be used to perform the task. Building a dialogue state tracking system within a voice assistant is not trivial. In some embodiments, a dialogue state tracking system involving one or more large language models can be implemented downstream of a natural language understanding system to produce the tracked intent and the one or more tracked entities. In some embodiments, a dialogue state tracking system involving one or more large language models can be implemented upstream of a natural language understanding system to produce rephrased natural language text, which is in turn processed by the natural language understanding system to produce the tracked intent and the one or more tracked entities.

The invention belongs to the technical field of AI large models, and particularly relates to an intelligent secretary autonomous learning and memory large model tool application which is used for understanding inquiry and commands of clients and achieving natural language inquiry, voice assistant functions and efficient client service interaction. According to historical behaviors, browsing records and purchase records of the user, a user portrait is generated, personalized commodity recommendation is provided, and the conversion rate is increased; the description, price and inventory information of commodities are managed and updated, and it is ensured that latest commodity data are displayed to users; the user is assisted in order placing and payment, and tracking, confirmation and exception handling tasks of the order are processed; the interaction history of the user is stored and analyzed to help formulate a marketing strategy and provide personalized customer support; users are helped to know commodity inventory conditions and logistics tracking information, and inventory management and distribution processes are optimized; and user feedback and product evaluation are collected and analyzed, so that the product quality and the user experience are improved, and the effect of intelligently supervising the e-commerce platform is achieved.

The invention discloses a multifunctional indoor lighting system based on AI, and provides an intelligent and personalized lighting scheme. The system comprises a human body detection module, a light induction module, an emotion recognition module, a music rhythm module, a central control module, a user interaction module and a communication module. The human body detection module uses an infrared sensor, a camera or a radar to monitor personnel activities in real time and automatically control a light switch. The light sensing module dynamically adjusts the brightness through a photosensitive sensor to ensure optimal illumination; the emotion recognition module analyzes emotion in combination with a camera and an AI algorithm, and automatically adjusts the color and brightness of light; the music rhythm module captures music rhythms through a microphone, so that light changes along with the music rhythms, and atmosphere is created. The central control module integrates sensor data, and utilizes an AI algorithm to accurately adjust the light effect. A user can set a personalized scene through a smart phone APP, a voice assistant or a touch panel. And the communication module adopts a Wi-Fi or 2.4 G wireless technology to realize equipment interconnection and remote control. The system has the advantages of intelligent control, energy conservation, environmental protection, multifunctional integration and the like, is widely suitable for families, offices and entertainment venues, and improves the lighting experience.

The embodiment of the invention provides a dialogue method and electronic equipment, and the method comprises the steps: displaying an interface of a main session in a first application, the main session being a session between a user and a first intelligent assistant provided by the first application; in response to a first operation of the user in the interface of the main session, displaying an interface of at least one independent session in the first application, the at least one independent session including a first independent session, the first independent session being a session between the user and a second intelligent assistant, the second intelligent assistant calls the intelligent dialogue service for the first application aiming at the first scene. According to the dialogue method and the electronic equipment provided by the embodiment of the invention, the use efficiency of the intelligent voice assistant can be improved.

Examples described herein relate to portable playback devices, such as smart headphones and earbuds, and ultra-portable devices having built-in voice assistants. Some example techniques relate to user interaction with voice assistants. Further example techniques relate to voice guidance played back by the headphones to guide the user under certain conditions.

An interface display method and a device are described. An original interface may be moved to reserve space for displaying an interface of a first application such as a voice assistant, so that the interface of the first application does not block the original interface. This helps a user interact with the original interface and the interface of the first application. For example, an electronic device displays a first interface. The electronic device detects an operation in which the user indicates to start a first application. The electronic device displays the first interface in a first area of a screen, and displays a second interface in a second area of the screen. The second interface is an interface of the first application, and the first area does not overlap the second area.

There is disclosed a computer-implemented method of training an interactive voice assistant (IVA), including prompting an artificial intelligence (AI) to assume a persona and to seek to fill a service request by a service provider that operates the IVA; receiving, from the AI, a query for the IVA; providing the query to the IVA; receiving, from the IVA, a response to the query; continuing to permit the AI to query the IVA until the AI is satisfied that the IVA has filled the service request or that the IVA is unable to fill the service request; logging the queries and responses between the AI and the IVA; and based on the logging, debugging or improving the IVA.

The disclosure relates to a method and apparatus for processing a voice input. The method includes identifying a first candidate text output from a first list of candidate text outputs, the first list of candidate text outputs being generated based on a voice input being received from a user. The method includes validating a feasibility of executing the first candidate text output based on an analysis of a first contextual data related to the first candidate text output, user-specific data, historical data related to previous validations of the voice assistant device, and one of an execution result or a no-action result corresponding to the first candidate text output. The method includes, based on the feasibility of executing the first candidate text output being validated as not feasible for execution, identifying failure information associated with the first candidate text output.

Systems and processes for providing personalized suggestions indicating that a task may be performed using a digital assistant of an electronic device are provided. An example method includes, at an electronic device with a display, receiving context data associated with the electronic device; determining, based on the context data, a task that may be performed by a digital assistant of the electronic device in response to a natural-language expression; determining, based on the context data, whether suggestion criteria associated with the determined task are satisfied; and in accordance with a determination that the suggestion criteria are satisfied, providing a suggestion indicating that the determined task may be performed using the digital assistant of the electronic device.

The invention provides a voice assistant interaction method and electronic equipment, the electronic equipment can determine whether to wake up a voice assistant according to own equipment posture and equipment action, and the equipment posture can be determined by combining measurement data of various sensors of the electronic equipment. For example, measurement data of an IMU, measurement data of an ultrasonic signal, touch information of a capacitive screen or measurement data of a proximity sensor and the like. The probability that the voice assistant is awakened by mistake is lower, and the energy utilization efficiency of the electronic equipment is higher. The pickup method and the feedback method executed by the electronic device can be determined according to the device posture of the electronic device, the voice interaction process of the user is more natural, and the usability of the voice assistant is better.

The present disclosure provides a voice interaction method and apparatus, an electronic device, a storage medium and a vehicle. The method comprises: acquiring a first voice generated in a first sound zone in a compartment in a sound zone locked mode, wherein the first sound zone is a sound zone for awakening a voice assistant; in response to the first voice triggering a multi-person interaction mode, acquiring a second voice generated in a second sound zone in the compartment, and determining a second intent of the second voice; and when determining that the second intent is correlated to a first intent, responding to the second intent, wherein the first intent is correlated to the first voice. When a sound zone is locked, an intent of a non-awakening user can be responded on the basis of the correlation between the second voice and the first voice, so that the technical effect of interaction between multiple persons and an in-vehicle infotainment system in a sound zone locked mode can be realized, thereby improving user experience.

The invention belongs to the technical field of speech recognition and acoustic modeling, and particularly relates to a high-fidelity end-to-end acoustic modeling method combining speech intelligibility reconstruction and autoregressive feedback optimization, which comprises the following steps of: firstly, establishing an intelligibility loss function based on a human speech intelligibility model; subjective definition features of the voice are reconstructed through a frequency spectrum reconstruction network, intelligibility related features are extracted and embedded in combination with a perceptual weighted filter bank, and the intelligibility related features and Mel frequency spectrum features are input into an acoustic encoder in parallel; based on this, by introducing a voice intelligibility reconstruction and autoregressive feedback mechanism, while the conciseness of an end-to-end voice modeling framework is maintained, systematic improvement of voice sharpness, naturalness and stability is realized, and the method can be widely applied to scenes such as intelligent voice assistants, voice transcription, virtual anchors, voice restoration, cross-language voice generation and the like. And the method has extremely high practical application value and popularization prospect.

The invention discloses a vehicle control method, device and equipment and a computer readable storage medium, and belongs to the technical field of vehicles. The method comprises the following steps: in response to a starting instruction for a voice assistant function configured in a vehicle, determining the running state of the vehicle and the remaining power percentage of the vehicle; under the condition that the running state of the vehicle meets the running requirement and the percentage of the remaining electric quantity of the vehicle is larger than a target threshold value, image acquisition equipment installed in the vehicle is controlled to be started; a first video collected by an image collection device in a first time period is obtained, the first video comprises a first gesture action of a driver and passengers of the vehicle, and the starting time of the first time period is the time for starting the image collection device; determining a first control instruction corresponding to the first gesture action; and controlling a control system corresponding to the first control instruction to execute a control operation corresponding to the first control instruction. According to the method, the use frequency of the voice assistant function is improved, and the vehicle control efficiency is higher.

A method comprising causing a voice assistant and a recommendation engine that are executing in an infotainment system of a vehicle to cooperate in processing a vehicle occupant's acceptance of a recommendation proposed by the recommendation engine by having an interface to enable the recommendation engine to provide recommendation context to the voice assistant to enable the voice assistant to resolve an ambiguity in the occupant's acceptance of the recommendation.

The invention relates to a voice assistant resource search test method and device, a platform, equipment and a medium, and the method comprises the steps: constructing a request body according to the resource information of any test case in a test case set, and transmitting the request body to a to-be-tested interface of a resource search end to request target resource path information, the test case set is pre-constructed based on a resource data table, and the test cases are literary and style format cases containing resource information; constructing a test message based on the resource information and the target resource path information, and sending the test message to at least one target test client in a notification mode to request to return information; and identifying target resource test information from the resource page screenshot according to the preloaded configuration file and the resource type in the return information, and determining a test result according to the target resource test information and a preset resource identifier in the return information.

The invention provides a voice interaction method and device, electronic equipment, a storage medium and a vehicle, and the method comprises the steps: obtaining a first voice generated by a first voice region in a carriage in a voice region locking mode, enabling the first voice region to be a voice region for waking up a voice assistant, and responding to the first voice to trigger a multi-person interaction mode, a second voice generated in a second voice area in the compartment is obtained, a second intention of the second voice is determined, the second intention is responded under the condition that it is judged that the second intention is related to the first intention, the first intention is related to the first voice, and the second intention can be responded when the voice area is locked. And the intention of the non-awakening user is responded according to the correlation between the second voice and the first voice, so that the technical effect of interaction between multiple persons and the vehicle-mounted terminal in the voice area locking mode can be realized, and the experience effect of the user is improved.

The invention relates to a voice wake-up method and device, electronic equipment, a storage medium and a vehicle. The method comprises the following steps: collecting a target voice signal; when it is detected that the target voice signal comprises the wake-up word and the generation position of the target voice signal is recognized to be in the cabin of the vehicle based on a position model, the current gear of the vehicle is determined; and awakening a voice assistant in the cabin under the condition that the target voice signal is determined to be the voice signal in the cabin based on the current gear. According to the method, the personnel outside the cabin of the vehicle can be prevented from waking up the voice assistant in the cabin, and the safety of the vehicle is improved.

The invention belongs to the technical field of artificial intelligence, and discloses a cerebral palsy child multi-mode rehabilitation training system based on artificial intelligence. The system comprises a data acquisition and preprocessing module, a multi-modal information fusion module, a personalized rehabilitation training suggestion generation module, an interaction and excitation module, a progress tracking and abnormity early warning module, a doctor and parent end APP management module and a system control and management module. According to the invention, when the interaction and excitation module and the multi-modal information fusion module are combined and applied to rehabilitation training of children with cerebral palsy, the training effect can be significantly improved, and accurate evaluation and personalized training can be realized; the child patient can be trained in a more real and attractive environment, meanwhile, the enthusiasm of the child patient is further stimulated through the gamification excitation design, and the training motivation of the child patient is enhanced through real-time feedback encouragement.

The invention relates to a vehicle-mounted voice assistant awakening method, device and system and a vehicle. The method comprises the following steps: in response to a contact operation for driving control equipment, a wake-up control module obtains a contact parameter corresponding to the contact operation, determines whether a vehicle-mounted voice assistant meets a wake-up condition according to the contact parameter, and generates a wake-up instruction for the vehicle-mounted voice assistant under the condition that the vehicle-mounted voice assistant meets the wake-up condition. By adopting the method, the awakening accuracy and portability of the vehicle-mounted voice assistant can be improved.

The disclosure relates to a method and apparatus for processing a voice input. The method includes identifying a first candidate text output from a first list of candidate text outputs, the first list of candidate text outputs being generated based on a voice input being received from a user. The method includes validating a feasibility of executing the first candidate text output based on an analysis of a first contextual data related to the first candidate text output, user-specific data, historical data related to previous validations of the voice assistant device, and one of an execution result or a no-action result corresponding to the first candidate text output. The method includes, based on the feasibility of executing the first candidate text output being validated as not feasible for execution, identifying failure information associated with the first candidate text output.

According to the translation device of the intelligent vehicle-mounted system and the control system of the translation device, the vehicle-mounted external translation device providing a multi-language translation service is constructed and generated through the service data terminal and the microphone, a driver and passengers are translated based on the NLP technology, and then a voice assistant of a vehicle-mounted sound control system is awakened by playing translation voice, so that the translation service of the vehicle-mounted system is improved. And the control instruction is issued to the voice assistant to complete vehicle control in a non-default language, so that the intelligent level of the vehicle is enhanced, a driver and passengers can realize part of functions of the vehicle through voice control even under the condition that the language is impassable, the driving safety is remarkably improved through games, and the driving experience is improved. In addition, the translation device can translate and play response voice of the voice assistant, intelligent interactive conversation between the driver and the vehicle-mounted sound control system is achieved, meanwhile, the translation device supports offline and networking work modes, and the voice translation task can be completed even under the condition that no network exists.

The invention discloses a somatosensory dance game method and device for generating a virtual character based on user timbre and a computer readable storage medium, and the method comprises the steps: activating a preset voice assistant to carry out interactive communication with a user when a virtual game character is created; collecting voice data when the user communicates with the voice assistant; generating a timbre model matched with the timbre of the user according to the voice data and a pre-trained machine learning model; and generating a voice and an image of a virtual character based on the tone model. The somatosensory dance game method for generating the virtual character based on the tone of the user has the advantage of intelligently and individually creating the virtual character.

Methods, computing systems, and technology for personalizing a user experience in a vehicle. For example, a computing system may be configured to receive a user prompt from a user, wherein the user prompt is input into a voice assistant on-board the vehicle. The computing system may be configured to process the user prompt with a master language model agent. The master language model agent may determine, based on environmental data, one or more intermediate actions responsive to the user prompt. The master language model agent may evaluate, based on the environmental data, whether each intermediate action of the one or more intermediate actions satisfies the user prompt. The computing system may be configured to output one or more command instructions to the voice assistant, wherein the one or more command instructions cause the voice assistant to provide, using one or more human-machine interfaces, a response to the user prompt.

The invention relates to a multi-modal emotion recognition fusion and communication method oriented to real-time human-computer interaction, and aims to solve the problems of asynchronous modal time sequence, non-uniform feature dimension, single fusion mode, high communication delay and the like of an existing emotion recognition system and improve the recognition precision and real-time performance of the system. According to the method, voice, images and physiological signals are synchronously collected through a multi-modal input module, feature alignment is achieved through time sequence interpolation, dynamic time warping and space coordinate mapping, double-domain feature fusion is conducted in combination with a time path network and a space path network, emotional state judgment is completed through a lightweight neural network, and the emotional state recognition accuracy is improved. And outputting six types of basic emotions and confidence coefficients. Meanwhile, a low-delay communication protocol based on UDP clipping extension realizes rapid feedback of emotion data, and in combination with modal priority scheduling, data compression and bandwidth sensing mechanisms, high-efficiency and low-delay transmission is ensured, and the real-time interaction requirement in a weak network environment is met. The method has the advantages of high accuracy, low power consumption, low time delay and flexible deployment, is suitable for various real-time interaction application scenes such as voice assistants, virtual customer service, emotion accompanying and telemedicine, and has wide application value and market prospect.

The invention relates to the technical field of speech synthesis, and particularly discloses a speech synthesis method and system for controllable latent variable modeling based on semantic distillation, and the method comprises the steps: converting a Mel spectrum into continuous latent variable distribution through a speech coding module, generating continuous latent variables through re-parameterization sampling, introducing a self-supervised model for semantic distillation, and carrying out the semantic distillation. According to the method, alignment of latent variables and semantic features is constrained through marginal cosine similarity and distance matrix structure loss, a text encoder maps a phoneme sequence into latent variable distribution, time sequence alignment of a text and the latent variables is achieved in combination with monotonic alignment search, and a decoder reconstructs the latent variables into a Mel spectrum. According to the method, waveform synthesis through a vocoder and total loss function joint optimization reconstruction, KL divergence, distillation, text alignment and confrontation loss are carried out, discrete information loss is avoided through continuous latent variable modeling, semantic consistency and text alignment efficiency are enhanced, the naturalness, coherence and real-time performance of synthesized voice are improved, and the method is suitable for scenes such as voice assistants and virtual anchors.