48 results about "False awakening" patented technology

The invention discloses a voice awakening performance improvement method and device. The voice awakening performance improvement method comprises the following steps of detecting an input signal in avoice frame form, and judging whether the voice frame is a suspected voice frame or a non-voice frame; determining an effective voice segment based on the determined suspected voice frame and the non-voice frame, and determining a weighting factor based on the ratio of the suspected voice frame and the non-voice frame in the input signal; inputting the effective voice section into an awakening model, wherein the awakening model is a multi-classification model obtained after supervised depth neural network learning is carried out on collected awakening word positive example samples and non-awakening word negative example samples; and acquiring the output of the adaptive awakening model, carrying out weighted calculation on the output by using the weighting factor, and determining whether the equipment is awakened or not based on output after weighing calculation. According to the scheme provided by the invention, the false awakening rate of a noise scene can be effectively inhibited, and further, the awakening rate for a specific user can be improved in a targeted manner.

The embodiment of the invention discloses a sound area amplitude acquisition method and device, an electronic device and a storage medium, wherein the method comprises the steps of collecting the voice data of a target sound area in real time, wherein the voice data comprises the audio signals corresponding to a plurality of sampling points; storing the audio signal by comparing the amplitude of the current audio signal to be stored with the amplitude of each stored audio signal, and selecting whether to store the current audio signal or not according to the comparison result; and calculatingthe current sound area amplitude of the target sound area according to the amplitudes of all the stored audio signals. According to the embodiment of the invention, the amplitude of the sound area canbe accurately obtained, so that the source sound area of the voice data is accurately positioned according to the amplitude of the sound area, the problem of false awakening caused by insufficient isolation of the four sound areas is solved, an amplitude data basis is provided for correcting false awakening of a voice engine, the calculation amount is small, and the system resources are saved.

The invention relates to a voice wake-up method. The method comprises the steps: receiving a plurality of voice wake-up messages sent by a plurality of electronic devices, wherein each voice wake-up message comprises a distance and a wake-up energy value, the distance represents the distance between the electronic devices sending the voice wake-up messages and a sound source, and the wake-up energy value represents the energy value of a wake-up word sent by the sound source and received by the electronic devices sending the voice wake-up messages; according to the distances and the wake-up energy values in the plurality of voice wake-up messages of the plurality of electronic devices, determining whether the energy attenuation of the wake-up word emitted by the sound source conforms to theattenuation rule of the sound energy radiated by a point sound source; and when it is determined that the energy attenuation of the wake-up word emitted by the sound source does not accord with the attenuation rule of the sound energy radiated by the point sound source, sending a wake-up forbidding instruction to the plurality of electronic devices. According to the scheme provided by the invention, the false awakening rate of the device can be reduced.

The embodiment of the invention provides a voice processing method and device, equipment and a storage medium, and aims to reduce the problem of false awakening. The method comprises the following steps: analyzing voice data to obtain one or more recognition keywords and corresponding confidence information; performing multi-level confidence verification on the one or more recognition keywords byadopting a plurality of confidence classifiers; and outputting the final keyword passing the multi-level verification. Through multi-level confidence verification, the accuracy of keyword recognitioncan be effectively improved, and the problem of false awakening is reduced.

The embodiment of the invention discloses a method and a device for outputting information. The method and the device can be used for an intelligent traffic scene. One specific embodiment of the method comprises the steps that: acquiring voice information within a preset time period before equipment is awakened, wherein an awakening model is arranged in the equipment and used for outputting presetresponse information when a preset awakening word is received; performing voice recognition on the voice information to obtain a recognition result; in response to determining that the recognition result does not include the preset awakening word, extracting feature information of the voice information; generating a counter-example training sample according to the feature information; and training the awakening model by using the counter-example training sample, and outputting the trained awakening model. According to the embodiment, the voice information during false awakening can be used asa counter-example sample to train the awakening model, additional corpus recording is not needed, the cost is saved, false awakening can be automatically inhibited, and the awakening accuracy is improved.

The invention discloses a wake-up word detection method, which comprises the following steps: obtaining speech information to be recognized, wherein the speech information to be recognized comprises atarget word set, the target word set is composed of at least two groups of repetitive target words, and each group of target words comprises at least one target word; acquiring a target speech feature vector according to the speech information to be recognized; using the speech recognition model for outputting the probability scores corresponding to each word in the speech information, wherein the target probability scores set includes the target probability scores corresponding to each target word. If the target probability score set satisfies the speech awakening condition, it is determinedthat the speech information to be recognized includes a target awakening word, and the target awakening word belongs to one awakening word in the awakening word set. The invention discloses a wake-upword detecting device and a terminal device. The invention improves the detection performance of the reduplicated awakening word, and has high awakening rate and low false awakening rate.

The invention provides a control method and an electronic device. The method comprises the steps of collecting a first input of the electronic device when the electronic device is in a first mode, wherein the first input is an input capable of switching the electronic device to be in a second mode from the first mode; obtaining first information which indicates spatial parameters of the electronic device; if the first information meets a first condition, switching the electronic device to be the second mode, wherein a quantity of inputs which the first mode can make a response to is smaller than a quantity of inputs which the second mode can make a response to; and if the first information meets a second condition, keeping the electronic device in the first mode. According to the control method and the electronic device, mode switching of the electronic device can be controlled based on spatial parameter related information of the electronic device, and the electronic device is controlled to make no response to the inputs and keep an original mode in the specific condition; and by applying the scheme, the problem of misoperation of the electronic device can be effectively solved, false awakening of a device screen and false startup of device applications are avoided, and the power consumption of the device is reduced.

The invention discloses a voice wake-up method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. The method comprises the steps: when a target audio is received, calculating the matching degree of the target audio and a wake-up audio; determining a threshold weight according to the user behavior data corresponding to the current time, thethreshold weight being less than or equal to 1, the user behavior data corresponding to the current time indicating that the higher the probability that the user performs the wake-up operation is, the smaller the threshold weight is; calculating a product of a preset threshold value and a threshold value weight to obtain a current threshold value; and if the matching degree is greater than the current threshold, executing a wake-up operation corresponding to the wake-up audio. According to the scheme, the threshold value is dynamically adjusted according to the user behavior data, the awakening rate is increased, and the false awakening rate is reduced.

The invention discloses a method for preventing false awakening of electronic tags. A 5.8G communication module is applied to the method, and comprises a 5.8G chip transmitting circuit, an awakening circuit, an RSSI (received signal strength indicator) detection circuit and a 5.8G chip receiving circuit. The RSSI detection circuit and the 5.8G chip receiving circuit are electrically connected in series. The RSSI detection circuit and the 5.8G chip transmitting circuit are electrically connected in series. The awakening circuit is electrically connected with an interruption detection module inseries. The interruption detection module is electrically connected with an awakening pulse period detection module in series. The awakening pulse period detection module is electrically connected with the RSSI detection circuit in series. The 5.8G chip receiving circuit is electrically connected with an RSSI sampling gap module in series. The method for preventing false awakening of the electronic tags has the advantages that start and stop of the RSSI detection module are controlled by the RSSI sampling gap module, low power consumption is achieved, and accordingly, stable equipment operation is realized.

The present application provides an audio processing method, comprising: collecting input data; responding to the input data in a manner that satisfies the first condition if the input data satisfyingthe first condition satisfies the second condition; if the input data satisfying the first condition does not satisfy the second condition, ignoring the input data satisfying the first condition. Byusing the method, whether the input data satisfying the first condition satisfies the second condition is determined, whether the input data is responded to in the manner of the first condition is determined, and two conditions are judged on the input data, so that the judgment accuracy is high, and false awakening is prevented.

The invention provides a wake-up model updating method. The method comprises the following steps of: separating a command word and a wake-up word in original voice data; determining the probability ofcommand words according to the language model, and deleting the command words with the probability lower than a first threshold value in the command words; and inputting the wake-up word into a wake-up model, determining a score output by the wake-up model, and deleting the wake-up word with the score lower than a second threshold value in the wake-up word. Therefore, the built wake-up model withhigher false wake-up probability is updated, command words and wake-up words are screened respectively, and the wake-up words are closer to preset wake-up words by eliminating false wake-up data or distortion data. Meanwhile, the command words serve as training data, the number of the training data is guaranteed, and the false awakening probability of the updated awakening model is lower.

The invention provides an air conditioner voice control method and device as well as an air conditioner. The air conditioner voice control method comprises the following steps of acquiring environmental parameters of a space where the air conditioner is located, wherein the environmental parameters at least comprise a voice command; judging a current scene of the air conditioner based on the environmental parameters; and reducing the voice recognition sensitivity of the air conditioner if the current scene of the air conditioner belongs to an anti-false awakening scene. According to the air conditioner voice control method and device as well as the air conditioner, risks of false awakening of the air conditioner can be reduced.

The invention discloses a voice wake-up method and device, electronic equipment and a storage medium, and relates to the technical field of voice processing, and the method comprises the steps: obtaining an input voice collected by an audio collector; matching the input voice based on a first voice matching model to obtain a first probability output, the first probability output being used for indicating a probability that the input voice contains a specified text; obtaining at least one probability output output by the first voice matching model before the current first probability output, and taking the probability output as a second probability output; fusing the first probability output and the second probability output to obtain updated first probability output; taking the updated first probability output as a first matching result of matching the input voice by a first voice matching model; and if the first matching result indicates that the input voice contains the specified text, waking up a terminal. According to the method, the historical output and the current output are fused, so that the keyword recognition accuracy can be improved, and the false awakening rate is reduced.

The invention discloses a data detection method and device for false awakening of equipment, a mobile terminal and a storage medium, and the method comprises the steps: obtaining the state informationof monitored equipment when there is voice, determining an awakening result of the monitored equipment according to the state information of the monitored equipment, and determining whether the monitored equipment is mistakenly awakened or not according to the awakening result and the voice. According to the data detection method for false awakening of the equipment, whether the monitored equipment is really awakened or not is analyzed from the state information of the monitored equipment, the awakening result of the monitored equipment is obtained through the state information after awakening, so that mistaken awakening detection data can be obtained according to the awakening result and the sent voice, a better basis is provided for adjusting and improving the success rate of voice awakening, and awakening operation is more accurate.

The embodiment of the invention discloses a radio frequency chip signal receiving method, device and system, and the method comprises the steps: carrying out the configuration in advance through a host program, and setting a sleep period and a self-awakening period of a radio frequency chip; when the elapsed dormancy time reaches a time threshold corresponding to the dormancy period, triggering the radio frequency chip to execute a self-awakening operation; after the radio frequency chip is automatically awakened, controlling the radio frequency chip to enter a receiving mode; detecting whether an effective receiving signal exists or not based on a preset state control module; if it is judged that the effective receiving signal exists, an interrupt signal is output to a host, and the hostis awakened to execute subsequent transaction processing operation. By adopting the method provided by the invention, the electric quantity waste caused by false awakening and asymmetry can be greatlyreduced, the success rate of segmented charging on a highway is conveniently improved, and the economic loss caused by missed fee deduction is reduced.

The embodiment of the invention provides a voice recognition method, a voice recognition device and electronic equipment, relates to the technical field of smart home, and can not only omit key pressing operation of a user, but also improve the problem of false awakening. The voice recognition method comprises the following steps: monitoring voice broadcast, and determining whether information ofthe voice broadcast comprises target voice information or not; when the information is confirmed to comprise the target voice information and an intelligent device is in the off-line state currently,converting the intelligent device into an awakening state from the off-line state; when monitoring that the voice broadcast is stopped, determining whether the information of the voice broadcast within a preset duration comprises a corresponding control command or not; and when the information is confirmed to comprise the control command, executing the control command.

The invention discloses a method and a device for reducing false awakening of intelligent voice equipment, and computer equipment. The method comprises the following steps: acquiring echo channel dataand microphone channel data of an intelligent voice equipment; processing the echo channel data and the microphone channel data to obtain an echo channel frequency domain signal and a microphone channel frequency domain signal; performing echo cancellation on the microphone channel frequency domain signal through an echo cancellation algorithm to obtain an echo-cancelled microphone channel frequency domain signal; calculating a first mutual interference coefficient of the microphone channel frequency domain signal after echo cancellation and the echo channel frequency domain signal; acquiringa specified data value corresponding to the microphone channel frequency domain signal after echo cancellation according to a preset rule; judging whether the first mutual interference coefficient isgreater than a preset first threshold value or not, and judging whether the specified data value is smaller than a preset second threshold value or not; and if yes, adjusting the wake-up threshold ofa wake-up word to a preset specified wake-up threshold. The possibility that the intelligent voice equipment is mistakenly awakened can be effectively reduced.

The invention discloses an end-to-end wake-up word detection method and device. The method comprises the steps of receiving to-be-detected voice; sequentially extracting acoustic features of each voice frame in the to-be-detected voice; inputting the extracted acoustic features into a pre-constructed acoustic model to obtain a posterior probability of a target pronunciation unit in each voice frame output by the acoustic model; taking each target pronunciation unit as a node, and inserting virtual mute nodes in front of and behind the target pronunciation unit to obtain a target-time relationship matrix; calculating the cumulative probability of each node in the target-time relation matrix frame by frame; determining an optimal path according to the cumulative probability of each node in the matrix; and determining a wake-up word detection result according to the optimal path. According to the invention, the accuracy of a detection result can be improved, and the false awakening rate is reduced.

The invention discloses a wake-up code Chirp signal generation method for realizing low power consumption and a communication method. The wake-up code Chirp signal generation method comprises the following steps: generating a fixed part; generating a circulation part capable of identifying different terminal devices based on the identification information of the terminal devices; and combining thefixed part and the cyclic part to form a wake-up code Chirp signal. The communication method comprises the wake-up code Chirp signal generation method; terminal equipment carries out CAD detection byutilizing the fixed part, and if the detection is passed, the terminal equipment is matched with the circulating part, and otherwise, the terminal equipment enters a sleep state; and the terminal equipment is matched with the circulation part, and if the matching is passed, the receiving state is kept to continue to receive the signal, otherwise, the terminal equipment enters the sleep state, andenergy saving is achieved. According to the wake-up code Chirp signal generation method, the to-be-awakened terminal equipment can be awakened uniquely and accurately, and the frequency of false awakening is effectively reduced, so that the energy consumption is reduced, and particularly, the power consumption of the wireless low-power-consumption system can be further reduced.

The invention belongs to the technical field of automobile cabins, and discloses a cabin system voice recognition test method, device and equipment and a storage medium. The method comprises the following steps: acquiring a preset audio signal, and sending the preset audio signal to a cabin system entertainment host; receiving an echo audio signal fed back by the cabin system entertainment host based on the preset audio signal; determining the number of echo signals and environmental noise data of the echo audio signals according to a preset analysis program; and determining a false wake-up rate of the cockpit system according to the number of the echo signals and the environmental noise data, and updating a voice database of the cockpit system according to the false wake-up rate. According to the invention, the voice interaction function test in the cockpit information entertainment host is satisfied through the preset audio signal, whether the voice wake-up is correct or not can be identified by analyzing the echo audio signal through the preset analysis program, and the false wake-up rate can be calculated by analyzing the number of the echo signals and the environmental noise data. And the accuracy of the performance test of the intelligent voice false awakening is improved.

The invention discloses an awakening performance optimization method and system based on mode switching. The method comprises the steps: acquiring awakening words; setting awakening modes, including asecondary awakening mode, a deep sleep mode and a common awakening mod, of to-be-awakened equipment according to an application scene; determining a secondary threshold value corresponding to the secondary awakening mode, a sleep threshold value corresponding to the deep sleep mode and a standard threshold value corresponding to a common awakening mode; and controlling the to-be-awakened equipment to switch among the three modes according to the awakening word, the secondary threshold, the sleep threshold and the standard threshold. Compared with a single common awakening mode, the awakeningperformance optimization method and system disclosed by the invention have the advantages that the secondary awakening mode is added, so that the awakening rate is improved; the deep sleep mode is added, so that the false awakening rate is reduced, and increasingly stringent performance indexes of a user can be met.

The embodiment of the invention discloses an awakening method and device of a vehicle-mounted voice system, a vehicle and a medium. The method comprises the steps that when audio data of a user are collected, whether an awakening word audio segment exists in the audio data or not is determined; if so, determining whether an adjacent audio segment exists in the wake-up word audio segment or not; and if not, awakening the vehicle-mounted voice system according to the awakening word audio segment. According to the embodiment of the invention, the false awakening rate of the vehicle-mounted voicesystem is reduced, and the awakening accuracy of the vehicle-mounted voice system is improved.

The invention provides a voice awakening method and electronic equipment. The method comprises the steps that a first electronic device collects a first voice signal in the located environment; if the audio is being played in the environment when the first voice signal is collected, the first electronic device obtains an audio signal corresponding to the audio in a wired or wireless communication mode; the first electronic equipment determines a first false awakening result according to the first voice signal and the audio signal; the first electronic equipment receives a second awakening result sent by the second electronic equipment; the first electronic equipment determines a third false awakening result according to the first false awakening result and the second false awakening result, and the third false awakening result is used for indicating whether awakening operation needs to be performed on to-be-awakened equipment in the local area network or not; and the first electronic device sends a third false awakening result to other electronic devices except the first electronic device in the local area network. The voice awakening method provided by the invention is helpful for improving the accuracy of false awakening judgment.

The invention relates to a smart television voice test technology, and discloses an automatic test method for the awakening rate and false awakening of a smart television. The method improves the awakening rate and the test efficiency and accuracy of false awakening. The power-on and power-off control of the smart television is realized by controlling a test tool through a control end, and no manual operation is required, so the test efficiency is improved; and the control end automatically sends an awakening word or a mistaken awakening corpus, serial port log printing of the smart televisionis monitored, the awakening times or the mistaken awakening times are counted by capturing the serial port printing information, and the power failure and the power on are controlled according to a program script, so that a fully-automatic test scheme is formed, human participation is avoided, the test efficiency is improved, and the test accuracy is certainly improved due to the formation of a unified test standard.

The invention discloses an awakening method and device of intelligent equipment, electronic equipment and a medium. The method comprises the steps of obtaining a voice signal collected by the intelligent equipment, and determining a wake-up probability that the voice signal contains a wake-up word; acquiring target feature information of the voice signal, inputting the target feature information into a pre-trained threshold model, and determining a target wake-up threshold corresponding to the target feature information; and if the wake-up probability of the voice signal is greater than the target wake-up threshold, waking up the intelligent equipment. According to the method, the awakening rate of the intelligent equipment is improved, and the false awakening rate is reduced, so that theuser experience is improved.

The invention provides an IoT chip-oriented multi-level voice intelligent awakening method and system. The method comprises the following steps: S1, when to-be-processed voice is acquired, executing S2; S2, judging whether the to-be-processed voice meets a first preset condition or not; when the to-be-processed voice meets the first preset condition, executing S3; otherwise, ending awakening; S3,judging whether the to-be-processed voice meeting the first preset condition meets a second preset condition or not; if yes, executing S4; otherwise, ending awakening; S4, judging whether the to-be-processed voice meeting the first preset condition and the second preset condition meets a third preset condition or not; if yes, executing S5; otherwise, ending awakening; and S5, comprehensively awakeing the IoT chip. According to the IoT chip-oriented multi-level voice intelligent awakening method provided by the invention, the awakening system is graded, and a step-by-step awakening and step-by-step power-on mechanism is adopted, so that the false awakening rate of the whole chip is reduced, and the overall power consumption of the chip is greatly reduced.

The invention discloses an intelligent equipment awakening method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring to-be-identified audio data, acquiring a first awakening result of the to-be-identified audio data according to a first false awakening algorithm, and acquiring a second awakening result of the to-be-identified audio data according to a second false awakening algorithm, and when the first awakening result and the second awakening result conform to a preset awakening relationship, awakening the intelligent equipment. Therefore, in the method provided by the embodiment of the invention, the first wake-up result and the second wake-up result of the to-be-identified audio data are obtained by using the first false wake-up algorithm and the second wake-up algorithm respectively, and whether to wake up the intelligent equipment is judged by combining the first wake-up result and the second wake-up result, so that the false wake-up rate of the intelligent equipment is reduced.

The invention discloses a connector awakened by the double touch sensors. The connector comprises two touch sensors arranged at the opposite sides of a shell, and a circuit board and a battery which are arranged in the shell, and a microprocessor is arranged on the circuit board and is connected with the battery and the two touch sensors. When the touch signals of the two touch sensors are detected at the same time, the microprocessor is awakened to a working state; and when the touch signals of the two touch sensors cannot be detected at the same time for a period of time, the microprocessorends the working state. Due to the fact that the two touch sensors are arranged back to back, the handheld habit of a normal person is well met. When a user holds the connector, the connector can be awakened without special actions so as to realize the non-inductive awakening. When the connector is placed, the two touch sensors which are oppositely arranged are not easily triggered at the same time, so that the false awakening is avoided, the energy consumption is reduced, and the service life of the battery is prolonged.

The invention discloses an awakening method and device and a terminal, which are applied to the technical field of artificial intelligence voice. In the scheme provided by the invention, the voice input by a user to equipment with an awakening function is monitored, and under the condition that the voice is monitored to contain an awakening word or under the condition that an awakening verification result is determined to possibly cause false awakening, a sensor on the equipment is started to detect whether someone exists around the equipment or not so as to assist in determining whether the current voice awakening is successful awakening or false awakening, false awakening is reduced, and the awakening accuracy is improved. According to the scheme provided by the invention, the problem that in the prior art, the wakeup rate is reduced when the wakeup threshold is improved due to reduction of the false wakeup rate is solved, and the problems that the cost of manpower, material resources, time and the like is increased and a new scene cannot be effectively and quickly adapted due to addition of a scheme of new model training to adapt to a new environment are also solved.

The invention provides an equipment awakening method and device, electronic equipment and a storage medium, wherein the method comprises the steps: obtaining audio data, converting the audio data into text data, and if it is determined that the text data contains a preset awakening word, determining a special degree value of the awakening word in the text data, wherein the special degree value is in positive correlation with the intensity of an awakening intention of a user for the equipment; and if it is determined that the special degree value belongs to the preset interval, awakening the device. Therefore, the device can be awakened only when the feature degree value of the awakening word belongs to the preset interval, and compared with an existing method of awakening as long as the awakening word exists, false awakening of the device can be effectively reduced, so that the user experience is improved.