25 results about "Acoustic voice analysis" patented technology

An object of the system according to the embodiment is to collect and analyze voices and sounds of children in a classroom and report the voices and sounds to a guardian.SOLUTION: A system includes a voice collection unit, an analysis unit, and a report unit. The voice collection unit collects voices or voices of children in the classroom. The analysis unit analyzes the voice data collected by the voice collection unit. The reporting unit reports a result analyzed by the analysis unit to the protector.SELECTED DRAWING: Figure 1

A gesture and voice-controlled interface device comprising one or a plurality of gesture sensors for sensing gestures of a user; one or a plurality of audio sensors for sensing sounds made by the user, and a processor configured to obtain one or a plurality of sensed gestures from said one or a plurality of gesture sensors and to obtain one or a plurality of sensed sounds from said one or a plurality of audio sensors, to analyze the sensed gesture and sensed sounds to identify an input from the user, and to generate an output signal corresponding to the input to a controlled device

An object of a system according to an embodiment is to enable a hearing-impaired person to recognize a siren sound of an emergency vehicle during driving and to drive safely.SOLUTION: A system according to an embodiment includes a siren sound analysis unit, a sign language video display unit, and an in-vehicle system cooperation unit. The siren sound analysis unit analyzes a siren sound. The sign language animation display part displays the sign language animation on the basis of the information of the siren sound analyzed by the siren sound analysis part. The on-vehicle system cooperation unit transmits information of the sign language animation displayed by the sign language animation display unit to the on-vehicle system.SELECTED DRAWING: Figure 1

The system according to the embodiment aims to analyze the sounds made by animals and understand their intentions and emotions. [Solution] A system according to an embodiment includes a collection unit, an analysis unit, and a generation unit. The collection unit collects animal sounds. The analysis unit analyzes the sounds collected by the collection unit to analyze the animal's intentions and emotions. The generation unit generates a response based on the analysis results obtained by the analysis unit.

The system according to this embodiment aims to enable users with poor articulation to communicate effectively using their own voice. [Solution] The system according to the embodiment comprises a collection unit, an analysis unit, a generation unit, a provision unit, and a training unit. The collection unit collects the user's voice. The analysis unit analyzes the voice collected by the collection unit and trains the generation AI. The generation unit performs reading aloud or transcription in its own voice based on the data analyzed by the analysis unit. The provision unit provides the voice and transcription generated by the generation unit to the user. The training unit performs articulation improvement training based on the data generated by the generation unit.

The present invention relates to the technical field of atmosphere lamp control, and discloses an audio analysis control method for a sound-collecting atmosphere lamp and an atmosphere lamp thereof, wherein the audio analysis control method for a sound-collecting atmosphere lamp comprises the atmosphere lamp collecting sound through a built-in sound receiving component, the sound collected by the sound receiving component includes music, the sound is analyzed and processed through audio analysis, the audio analysis comprises pre-processing the sound, and extracting the energy proportion of each frequency band of the sound by extracting the time domain features and the frequency domain features of the pre-processed sound, each of which includes a high-energy frequency band and a low-energy frequency band, and performing a moving average filter on the energy of the low-frequency band to generate an energy envelope. The present invention processes the collected music signal in real time through audio analysis to extract information such as the rhythm, frequency and intensity of the music. Audio analysis technology ensures that the lights can be highly synchronized with the music, thereby improving the user experience.

The system according to the embodiment aims to analyze the voice of an infant and facilitate smooth communication between family members. [Solution] A system according to an embodiment includes a collection unit, an analysis unit, an output unit, a display unit, an audio output unit, and a learning unit. The collection unit collects the voice of an infant. The analysis unit analyzes the voice collected by the collection unit. The output unit outputs the results of the analysis by the analysis unit. The display unit displays the results of the analysis by the analysis unit. The audio output unit outputs the results of the analysis by the analysis unit as audio. The learning unit learns new voice collected by the collection unit.

Provided are an electronic device for analyzing and searching for a photo on the basis of a sound, and an operation method therefor. The electronic device may: acquire a sound signal by means of a microphone for a preset time before and after a time point at which a photo is captured by using a camera; input a feature vector extracted from the sound signal to an artificial neural network model; acquire at least one type of emotion or taste pertaining to the sound signal by analyzing the feature vector by using the artificial neural network model; and store the at least one type of emotion or taste as metadata for the photo captured by using the camera.

The application relates to the technical field of in-vehicle interaction, and provides an in-vehicle emotion pacification method and device, electronic equipment and a vehicle. The method comprises the following steps: acquiring in-vehicle sound; analyzing the in-vehicle sound to obtain an analysis result, wherein the analysis result comprises passenger information and passenger emotion of in-vehicle passengers; when the passenger emotion satisfies a pacification condition, determining a target interactive object of the in-vehicle passengers according to the passenger information; generating a replica sound of the target interactive object, playing pacification audio comprising the replica sound to the in-vehicle passengers, and pacifying the in-vehicle passengers by using the pacification audio. Since the pacification audio comprises the replica sound of the target interactive object, the in-vehicle passengers are more intimate and trust the target interactive object, are easy to stabilize the emotion after hearing the targeted pacification audio comprising the replica sound, the pacification effect is more ideal, the child passengers are more likely to stop crying, the driver's concerns are eliminated, the driving risks when there are child passengers are eliminated, and the problem that the pacification effect on the child passengers in the prior art is not ideal is solved.

The invention discloses a thyroid cancer postoperative care method based on multi-dimensional sound analysis, and belongs to the technical field of medical care. Comprising the following steps: step 1, constructing a composite feature vector covering pitch stability, a volume dynamic range, tone change, intonation emotion intensity and resonant cavity coordination; step 2, realizing individualized sound modeling and dynamic reference tracking so as to accurately describe postoperative sound changes; step 3, constructing an adaptive learning model capable of accurately capturing subtle changes of individual sounds; 4, an intelligent early warning mechanism is established, the abnormal vocal cord function is recognized in time, and the sensitivity and specificity of postoperative rehabilitation monitoring are improved; 5, sound feature analysis and clinical rehabilitation data are combined, personalized rehabilitation suggestions are generated, and intelligent optimization of a postoperative rehabilitation scheme is achieved; and step 6, constructing a multi-dimensional sound feature big data platform, and realizing safe storage and intelligent analysis of sound data.

A device based on artificial intelligence and a hybrid system for signaling the health status of a single man or woman, or if he or she has fainted or fallen into a coma, to doctors, police, and other competent persons and authorities, characterized in that this device consists of sensors that detect whether there are signs of life from him or her in the room, a clock, time relays, a telephone system, and all or part of any matter and instruments such as switches, toggles, reset switches, buttons, sensors, ultrasonic or infrared or electromagnetic wave sensors, transmitters and receivers, photodiodes, light diodes, twilight switches, light barrier reflections, optosensors, motion detectors, logic, software, digital, hardware, keyboards, signal generators such as sirens, warning lights, announcement generators such as in answering machines, video cameras, digital cameras, smoke detectors, fire alarms, voice recording devices, and microphones.Sound analyzer, voice analyzer, which can recognize various keywords and key phrases such as "screaming," "help," "ah," "loud coughing like in asthma," and so on, indicating pain or fear; navigation systems to know one's own location; central processing unit (CPU) operating system, memory, registers, arithmetic logic unit (ALU), modem, ISDN or similar devices, mouse, keyboard, and other similar interfaces and peripherals, and some or all sensors such as distance sensors using magnetic, infrared, or ultrasonic waves, motion detectors, light detectors, photodiodes, light transistors, limit switches, flow sensors, fire detectors, smoke detectors, microphones, limit switches, seat contacts, limit switches, switches that are activated by opening or closing doors and change their state from closed to open and vice versa, contacts that sensors,These devices, which generate signals based on the sound of breaking glass, include glass break detectors, hybrid systems, and so on. They process signals from sensors and measure the time since the last sign of life from the person. If this time exceeds a certain limit, the device determines whether the person is unconscious or not. If there are no clear signs of life, the device sends a call, written text, and / or spoken message, one or more times and in stages, to doctors, police, and other relevant persons and agencies, so that doctors and others can go to the person to help or rescue them.

A patient monitoring system includes a microphone that collects audio data from a patient. The audio data is used to generate audio characteristics to classify the audio data and analyze the audio data to determine a patient health status. The audio characteristics and the patient health status are tracked over time and used to monitor a patient respiratory condition. Based on a comparison between the audio characteristics and a database of audio characteristics associated with recorded health statuses of various patients, the system determines the patient health status. Based on the database of audio characteristics and associated health statuses, the system generates a report of the current patient health status.

We propose a substitute model for analyzing new detection targets. [Solution] The substitute model proposal device comprises: an acquisition unit that acquires an audio signal in which a sound containing the target sound to be detected is recorded; an analysis unit that generates multiple frames by dividing the audio signal into predetermined intervals and analyzes the distance between the deep features of the multiple trained models and the deep features of the frame sounds included in the multiple frames for each frame using multiple trained models that detect sounds different from the target sound; and a proposal unit that determines and outputs one or more substitute models from the multiple trained models to be used for detecting the target sound based on the analyzed distance.

The invention provides a method and system for identifying the state of equipment in an ultrahigh-voltage substation cabinet, and the method comprises the steps: processing collected image data, sound data and electrical parameter data, and inputting corresponding feature extraction and analysis models respectively to generate analysis results; and inputting the image analysis result, the sound analysis result and the electrical parameter analysis result into an equipment state recognition model constructed by adopting a neural network algorithm to obtain an equipment state recognition result. According to the method and the system, through an advanced image recognition technology and a machine learning algorithm, real-time automatic monitoring, accurate recognition and early warning of the state of the equipment in the cabinet are realized, manual frequent inspection is not needed, the safety risk caused by opening the cabinet by operation and maintenance personnel during manual inspection is avoided, the personal safety of the operation and maintenance personnel is guaranteed, and the working efficiency is improved. The operation and maintenance efficiency and safety of the transformer substation and the intelligent level of operation and maintenance management of the transformer substation are greatly improved.

The system according to the embodiment aims to provide a personalized training plan based on the characteristics of a user's voice. [Solution] A system according to an embodiment includes a recording unit, an analysis unit, a creation unit, a monitoring unit, and a feedback unit. The recording unit records the user's voice. The analysis unit analyzes the data recorded by the recording unit and extracts features of the user's voice. The creation unit creates an individual training plan based on the features extracted by the analysis unit. The monitoring unit monitors the user's practice in accordance with the training plan created by the creation unit. The feedback unit provides feedback based on the progress monitored by the monitoring unit.

An automatic inspection system includes a wireless slave device and a wireless master device. The wireless slave device includes a sound collection unit that collects sound generated from an inspection target object, an analysis unit that analyzes the collected sound to obtain a degree of difference between the collected sound and normal sound learned in advance and sound state information of the collected sound, as an analysis result, a wireless communication unit that wirelessly transmits data including the analysis results to a wireless master device, and a power supply unit that supplies power to the sound collection unit, the analysis unit, and the wireless communication unit. The wireless master device is configured to perform processing of receiving and managing the data from the wireless slave device, and transmitting the analysis result extracted from the data to a monitoring terminal that monitors a state of the inspection target object.

The invention relates to the technical field of sound analysis, in particular to a large pump station detection system based on forward and reverse voiceprint characteristics. The voiceprint extraction module is used for acquiring a sound source signal of a pump station unit and generating a voiceprint set and a current voiceprint feature; the voiceprint reconstruction module is used for inputting the current voiceprint features into a voiceprint reconstruction model for reconstruction mapping and outputting reconstructed voiceprint features; the noise stabilization module is used for calculating a mapping difference value between the current voiceprint feature and the reconstructed voiceprint feature, generating a dynamic residual error sequence through time sequence splicing, executing fractional order difference operation and outputting a stable residual error sequence; the voiceprint detection module is used for executing probability distribution fitting from the stationary residual sequence to obtain a Pareto model; and generating an abnormal signal and a voiceprint fault based on a Pareto model. According to the method, standard residual input for eliminating non-stationary interference is provided for extreme value theoretical probability fitting at the rear end through multi-dimensional feature extraction and fractional order difference processing at the front end, and the accuracy of early warning triggering is improved from the judgment dimension.

The invention provides a Bluetooth earphone with an intelligent noise reduction function, and the earphone comprises a master control Bluetooth chip which is responsible for processing all wireless communication and data transmission; the environment sound analysis module is provided with a high-sensitivity microphone array and is used for capturing and analyzing sound characteristics of the surrounding environment in real time; and the noise reduction adjusting module works in cooperation with the environment sound analysis module, and automatically adjusts the noise reduction level according to the analysis result so as to generate anti-phase sound waves counteracting the environment noise. According to the intelligent noise reduction Bluetooth earphone, the environment sound analysis module and the noise reduction adjustment module are integrated, the environment sound analysis module is provided with a high-sensitivity microphone array, and the noise reduction adjustment module dynamically adjusts the noise reduction level according to information and generates anti-phase sound waves counteracting the environment noise, so that the noise is actively neutralized; the active noise reduction technology is combined with the traditional physical isolation of external sound, so that a user can enjoy a clear and high-quality audio effect even in a noisy environment.

The invention provides an earphone sound analysis method and system and an electronic device, and the method comprises the steps: obtaining layered buffer audio data, carrying out the deep heterogeneous feature extraction and fusion, obtaining an original audio content fingerprint sequence, collecting a target feature vector sequence through a sliding window, and obtaining a target feature vector sequence; obtaining a content type label, a sound field scene label and a state vector by adopting a feature mixing model, and generating an audio optimization strategy through an audio optimization multi-agent model based on the content type label, the sound field scene label and the state vector; the audio optimization multi-agent model comprises a voice definition agent, a fidelity agent, a space sense agent, a comfort level agent, a noise reduction agent and an arbitration agent, executes an audio optimization strategy, processes layered buffer audio data through dynamic parameter tracking, outputs an optimized audio signal, and outputs the optimized audio signal. According to the invention, the adaptive optimization capability and cooperative adjustment capability of the earphone on the audio processing parameters in a complex scene are improved.

The system according to the embodiment aims to provide an assistant function that utilizes voice recognition in a moving space. [Solution] The system according to the embodiment includes a voice recognition unit, an analysis unit, a route guidance unit, an application operation unit, an API collaboration unit, and a suggestion unit. The voice recognition unit recognizes the user's voice. The analysis unit analyzes the voice recognized by the voice recognition unit. The route guidance unit provides route guidance based on the information analyzed by the analysis unit. The application operation unit operates smartphone applications based on the information analyzed by the analysis unit. The API collaboration unit acquires external information based on the information analyzed by the analysis unit. The suggestion unit makes specific suggestions to the user based on the information analyzed by the analysis unit.

The present invention relates to a sound source analysis device and, more particularly, to a sound source analysis device for analyzing a sound source by using a plurality of microphones. The sound source analysis device according to an embodiment of the present invention comprises: a speaker for outputting an alarm; a voice reception unit for receiving a voice; a housing that provides a first accommodation space for accommodating the speaker and a second accommodation space for accommodating the voice reception unit; and a voice cover coupled to the housing and sealing the voice reception unit from the outside, wherein the voice reception unit includes a plurality of microphone modules.

The invention relates to the technical field of sound analysis and the like, and provides a visual conference volume testing method and system, and the method comprises the steps: configuring an audio conference system comprising a plurality of microphones and loudspeakers, and forming an audio input and output pair; through the conference scene interaction interface, remote touch control is carried out to turn off the empty seat equipment, and any attendance position loudspeaker is selected as a test sound source; controlling the loudspeaker to simultaneously send preset test sound signals to all the attending position microphones; after each microphone receives the signal, the sound analysis unit calculates the test volume value of the position where the microphone is located, and sends the test volume value to the DSP processor; and the DSP processor operates a sound mixing algorithm based on the numerical values of all the microphones and a preset acoustic model, and calculates volume data of the horns corresponding to the microphones. According to the invention, rapid equipment configuration and global linkage test are realized by means of a visual interface, the efficiency is remarkably improved, data are acquired in a real sound field, the limitation of traditional isolated test is overcome, and accurate and repeatable system-level acoustic equalization is realized through objective data analysis.

The invention relates to the technical field of data analysis and processing, and further relates to a multichannel sound analysis system and method based on sound sensing, and the system comprises a scene controller which is configured to be used for building at least one standardized acoustic scene in an acoustic experiment space; the auditory meatus sensing module is used for collecting auditory meatus sound pressure signals; the reference sensing module is used for acquiring a reference sound pressure signal; the leakage driving module is used for switching between at least two different leakage states in a preset period; and the algorithm execution module is used for processing the acquired sound pressure signal of the ear canal and the reference sound pressure signal, and outputting a sound analysis curve by jointly analyzing the switching period of the sound pressure signal and the leakage state. According to the invention, objective quantification and cross-device comparison of noise reduction performance are realized.