31results about How to "Improve wake-up effect" patented technology

The invention provides an intelligent trip method and system, wherein the method includes the steps: obtaining location information of a starting point and a destination, and carrying out route planning, to obtain a trip route and a trip mode; according to the location information of the starting point and the destination, calculating the distance between the starting point and the destination; obtaining the expected arrival time information and date; obtaining public transport means data information; according to the distance between the starting point and the destination, the public transport means data information, the time information and the date, calculating specific driving time; according to the expected arrival time and the specific driving time, calculating actual bell ringing time; and setting the bell ringing time of an alarm clock. Due to the adoption of the alarm clock, a user can set the alarm clock for reminding the user to start in advance according to user needs.

The application discloses a judgment method and device of voice wakeup, and electronic equipment. The method comprises the following steps: acquiring input voice, wherein a first model judges whetherto wake up according to the input voice; if the judgment result is to wakeup, returning to the wakeup length, and determining a wakeup voice segment; inputting a spectrogram of the wakeup voice segment to obtain a wakeup judgment result, wherein the second model is a dichotomy network model obtained according to the spectrogram training of a large amount of wakeup voice segments. Through the judgment method and device of voice wakeup, and the electronic equipment disclosed by the application, the wakeup voice segment is intercepted when the first judgment result is wakeup, and then whether towake up is further judged according to the spectrogram of the wakeup voice segment, the two-layer judgment is performed in the process, and the second judgment depends on the direct and effective wakeup voice segment, the mistaken wakeup caused by a certain environment elements can be eliminated, and the wakeup performance of the equipment is improved.

The invention discloses a wake-up model generation method, an intelligent terminal wake-up method and device. The invention discloses a wake-up model generation method, belongs to the technical fieldof voice wake-up, and the wake-up model generation method comprises the following steps: marking the start and stop time of each wake-up word contained in wake-up word audios in a sample audio set, and obtaining marked wake-up word audios, and the time length of the wake-up word audios is not fixed; adding noise to the marked wake-up word audio by using a negative sample audio containing background noise to obtain a positive sample audio; respectively extracting a plurality of audio frame features from the positive sample audio and the negative sample audio, and performing frame label labelingon the positive sample audio and the negative sample audio to obtain a plurality of audio training samples; and training the recurrent neural network by using the plurality of audio training samplesto generate a wake-up model. According to the embodiment of the invention, the model training is carried out by using the variable-length input recurrent neural network, so that the operation of manually intercepting the sample can be avoided, and the wake-up effect of the intelligent terminal is improved.

The invention discloses a noise reduction method and device for a vehicle-mounted environment, electronic equipment and a storage medium, whichcan be applied to the technical field of computers, in particular to the field of audio processing. The specific implementation scheme is as follows: an interference signal in a vehicle-mounted environment and a sound signal in the vehicle-mounted environment are received, wherein the interference signal comprises a vibration signal of a vehicle; noise reduction processing on the sound signal in the vehicle-mounted environment is carried out to obtain anoise reduction signal, wherein the noise reduction processing comprises the step of eliminating the interference signal from the sound signal in the vehicle-mounted environment. According to the embodiment of the invention, the vibration signal of the vehicle can be eliminated from the sound signal in the vehicle-mounted environment, thereby improving the noise reduction capability of echo elimination in the voice recognition system, improving the signal-to-noise ratio of an effective signal, and further effectively improving the voice recognition and awakening capability of the system.

The invention provides a wake-up word estimation method and device, a storage medium and an electronic device. The method comprises: acquiring words to be estimated, input by a user; extracting estimation characteristics of the words to be estimated, wherein the estimation characteristics are used for indicating distinctiveness of the words to be estimated, in acoustic and / or semantic aspect; using the estimation characteristics of the words to be estimated as input, processing with a pre-constructed wake-up word estimation model, and determining whether the words to be estimated are suitablefor serving as wake-up words. The wake-up word estimation method and device, the storage medium and the electronic device help improve the accuracy of wake-up word estimation results, thereby improving wake-up effect of wake-up words set by a user.

The invention discloses a control method for an air conditioner. The method includes the steps that after the air conditioner is controlled to play a quarter bell or a quarter bell audio frequency signal is detected, whether the quarter bell is closed or not is detected; and when it is not detected that the quarter bell is closed within the first preset time interval, running parameters of the airconditioner are adjusted so as to adjust the somatosensory temperature of a user. The invention further discloses a control detection device for the air conditioner and a computer readable storage medium. According to the control method for the air conditioner, by adjusting the somatosensory temperature of the user, the user is made to fell a sudden change so that the user can be awaken, and theawaking effect of the quarter bell is improved.

The invention discloses a self-defined awakening method and device applied to intelligent voice equipment. The method comprises the steps of receiving a registration text set by a user and collectingregistration voice of the user; computing a registration awakening threshold based on the registration text; computing a registration awakening score of the registration voice based on a universal awakening model, and judging that whether the registration awakening score is more than the registration awakening threshold or not; if the registration awakening score is more than or equal to the registration awakening threshold, aligning the registration voice with the registration text, and extracting a voice segment corresponding to each character in the registration text; and extracting a Gaussian posterior feature of the voice segment corresponding to each character, and generating a feature template of the whole registration voice based on the Gaussian posterior feature of each voice segment. According to the scheme provided by the method and the device, the quality of input voice can be optimized in a registration stage, so that the system can be much finer in a score computing module, and finally, a mistakenly awakening rate of similar words also can be reduced while an awakening rate is improved.

The invention discloses sleep conditioning equipment. The sleep conditioning equipment comprises a perfume spray device, a control device and a luminescent source, wherein the perfume spray device releases a sleep conditioning fragrance; the control device is electrically connected with the perfume spray device and controls the perfume spray device to release the sleep conditioning fragrance; theluminescent source is electrically connected with the control device, and the control device controls the luminescent source to emit sleep conditioning light for conditioning sleep after the sleep conditioning fragrance is released for a preset time. The sleep conditioning equipment has the advantages that the structure is simple and compact, the efficacy loss of essential oil molecules is effectively avoided, running is stable, the sleep conditioning equipment has sleep conditioning functions of helping sleep and waking, the sleep conditioning effect is good, carrying is convenient, the energy is saved, and the environment is protected.

The invention provides an alarm clock awakening method and a terminal. The alarm clock awakening method is applied to the terminal. The method comprises the steps of: when the terminal plays an awakening alarm, if the operation that a user stops the awakening alarm is received, suspending playing of the awakening alarm, and furthermore, obtaining the time when playing of the awakening alarm is suspended; broadcasting the time when playing of the awakening alarm is suspended, and determining the time when playing of the awakening alarm is suspended as the beginning time of a pre-set time slot; and, in the pre-set time slot, if real-time audios input by the user are received, judging whether the real-time audios are matched with pre-set standard audios or not, and, if the real-time audios are matched with the pre-set standard audios, stopping the awakening alarm. Thereby, the complex degree of the awakening alarm stopping operation is increased, so that the awakening effect is improved.

The invention discloses an equipment wakeup related system, method, device and equipment. The equipment wake-up method comprises the following steps: determining a wake-up word probability of a previous voice frame; adjusting a filtering coefficient according to the wakeup word probability and the previous voice frame; performing voice enhancement processing on the current voice frame according to the adjusted filter coefficient so as to suppress a human voice noise outside the target sound source in the current voice frame; and recognizing a wakeup word according to the current voice frame after the human voice noise is suppressed so as to adjust the equipment to a wakeup state. By adopting the processing mode, the wakeup feedback of the previous voice frame is combined, and the coefficients of the beam forming filter are updated so that the target voice and the human voice noise can be distinguished, and a reliable noise reduction effect is obtained; and therefore, the wakeup performance of the equipment under highly noisy human voice interference can be effectively improved.

The invention discloses a morning awakening robot. The morning awakening robot comprises a mounting plate; a mounting hole is formed in the mounting plate; a support plate is fixedly welded on one side surface of the mounting plate; a control box is arranged on the upper surface of the support plate; a controller is arranged in the control box; a first servo electric cylinder is fixedly mounted onthe lower surface of the support plate, and is electrically connected with the controller through a lead; a lamp box is fixedly welded on an output end of the first servo electric cylinder; an electronic alarm clock is mounted on the right side of the upper surface of the lamp box through a bracket and a rotating shaft, and is electrically connected with the controller through a lead; an LED lampcontrol mechanism is arranged in the lamp box; a stop switch is arranged on the lamp box, and is electrically connected with the controller through a lead; and at least three sets of LED lamps are fixedly mounted on a base plate of the lamp box. The morning awakening robot is safe and effective, awakens users through change of light rays, and achieves a good awakening effect.

The invention discloses an intelligent alarm clock and lamp control system based on sleep monitoring. The system comprises an intelligent wearable device worn on the body of a user, an intelligent alarm clock, an intelligent switch, an intelligent timer, mobile phone software and a lamp connected with the intelligent switch. The intelligent wearable device, the intelligent alarm clock and the intelligent switch are connected with the mobile phone software. According to the invention, by monitoring the sleep state of a user, an awakening effect can be achieved through the sleep awakening alarmclock, and the effect of automatically turning on and turning off the lamp can be achieved by matching with a light-operated switch; meanwhile, through sleep depth monitoring in cooperation with the ringing intensity, the range of applicable people can be improved, and the system is low in noise, good in awakening effect and high in comfort level; through the storage function of an intelligent timer, a sleep quality monitoring effect can be achieved, and the body health is benefited; and the whole process is high in intelligence and does not need manual operation, and the system is good in experience effect, high in functionality and wide in application range.

The invention discloses bamboo shoot fiber powder noodles. The noodles is prepared from the following raw materials in parts by weight: 60-70 parts of flour, 15-20 parts of bamboo shoot fiber powder, 10-12 parts of vital gluten, 4-6 parts of egg liquid, 4-6 parts of white sugar, 2-4 parts of white vinegar, 1-3 parts of baking soda, and 1-3 parts of edible salt. A preparation method of the bamboo shoot fiber powder noodles includes the following steps of step 1, preparing the raw materials; step 2, conducting dough kneading; step 3, conducting dough standing; step 4, conducting rolling; step 5, conducting pressing; step 6, conducting drying; and step 7, sterilizing the dried noodles in batches and packaging into bags. Bamboo shoot fiber powder is added into flour, so that the noodles are fine and smooth in mouth feel and fragrant in smell, and the content of soluble dietary fibers, the water-holding capacity and the swelling force are all improved differently; and after the vital gluten is added, wet gluten with a network structure is formed after the vital gluten absorbs water, and the wet gluten has good viscoelasticity, extensibility, thermal coagulability, emulsibility and film formability, so that the elasticity of the noodles is greatly improved.

The invention provides a voice wake-up method and device and a training method and device of a voice wake-up module. The voice wake-up method comprises the steps that a to-be-recognized voice signal is acquired, and the to-be-recognized voice signal comprises a wake-up word; performing wake-up word recognition on the to-be-recognized voice signal by using an acoustic model to obtain a wake-up word recognition result, the acoustic model comprising a whole word node, and the whole word node taking a word vector of a wake-up word as a weight; and performing voice wake-up according to the wake-up word recognition result, thereby improving the wake-up effect.

The wearable lower limb rehabilitation training apparatus comprises a first movement assembly, the first movement assembly comprises a first wearing part and a first sensor, and the first sensor corresponds to thigh muscles; the second movement assembly is rotationally connected with the first movement assembly and comprises a second wearing part and a second sensor, and the second sensor corresponds to the calf muscle; the pedal piece is fixed to the foot and is in universal connection with the second movement assembly; the first damper is electrically connected with the first sensor, the second damper is electrically connected with the second sensor, the first damper controls the rotation resistance of the first joint, and the second damper controls the rotation resistance of the second joint; when the training apparatus is in the cooperative motion state, the first driving part drives the first joint to rotate according to the rotating direction of the second joint. The device is small in overall size, convenient to carry and convenient to use, and has a good training effect in different rehabilitation stages of a patient.

This application relates to the field of artificial intelligence, and discloses a training method for awakening models, including: extracting audio frames from specified speech sentences in the training set to obtain an acoustic feature matrix; inputting the acoustic feature matrix into the keyword detector of the first model to obtain the first space feature, input the acoustic feature matrix into the encoder of the second model to obtain the second spatial feature, the first model is the wake-up model to be trained, and the second model is the trained noise reduction model; calculate the first spatial feature and the second spatial feature The difference of the spatial characteristics of the feature; according to the calculation method of the difference of the spatial characteristics corresponding to the specified speech sentence, calculate the difference of the spatial characteristics corresponding to all the speech sentences in the training set; Set the cross-entropy loss to form the loss function of training the wake-up model to train the wake-up model. The feature vector of the high-dimensional space is used as a knowledge distillation sample to assist in training the wake-up model and improve the wake-up effect.

The invention provides an air-conditioner indoor unit and a control method thereof. An air injection device is arranged at the edge of a machine shell of the indoor unit. The control method comprises the steps that an automatic awakening instruction of the air-conditioner indoor unit is obtained; environment parameters of a working environment of the indoor unit and / or state parameters of a target user are / is obtained; a working mode of the air injection device is determined according to the environment parameters and / or the state parameters; and the air injection device is driven to emit an impact airflow to the target user according to the working mode. Therefore, the air-conditioner indoor unit has the awakening function, the impact airflow adaptive to the environment parameters and / or the state parameters of the target user can be emitted to the working environment so as to awaken a user in the working environment, different awakening effects can be achieved, the user experience is improved, and the intelligent degree of the air-conditioner indoor unit is enhanced.

The present application discloses a method, device and electronic device for judging wake-up by voice, including: acquiring input voice, and a first model judges whether to wake up according to the input voice; if the judgment result is wake-up, returning the wake-up length to determine the wake-up voice segment ; Input the spectrogram of the wake-up speech segment into a second model to obtain a wake-up judgment result, wherein the second model is a two-class network model trained according to a large number of spectrograms of the wake-up speech segment. In the method, device and electronic device for judging wake-up by voice, when the result of the first judgment is wake-up, the wake-up voice segment is intercepted, and the wake-up voice segment is further judged according to the spectrogram of the wake-up voice segment. And the second judgment is based on the direct and effective wake-up voice segment, so it can eliminate false wake-up that may be caused by some environmental factors, and improve the wake-up performance of the device.

The invention discloses a dough waking-up device convenient for temperature control, and belongs to the technical field of dough making. The dough waking-up device comprises a dough waking-up box, locking universal wheels are fixedly mounted at four corners of the lower surface of the dough waking-up box, a box door is rotatably connected to the front end face of the dough waking-up box through hinges, a plurality of ventilation grooves are formed in the inner wall of the box door, a temperature control mechanism is connected to the box door, and a temperature adjusting box is fixedly connected to the inner end face of the dough waking-up box; a plurality of electric heating pipes which are uniformly distributed and are used for heating the dough waking-up box are mounted between the upper inner wall and the lower inner wall of the temperature adjusting box, the electric heating pipes are electrically connected with the temperature control mechanism, and the outer side of the temperature adjusting box is fixedly connected with a protective net which is matched with the temperature adjusting box; and a plurality of U-shaped plates which are uniformly distributed from top to bottom are fixedly connected to the inner walls of the two sides of the dough waking-up box. The dough waking-up device is provided with the temperature control mechanism, so that the temperature in the dough waking-up box can be conveniently regulated and controlled in time, and the dough waking-up effect of dough is improved.

The invention provides a data processing method applied to an electronic device. The method comprises the steps that a plurality of pieces of first historical voice data are acquired; at least one piece of first target voice data in the multiple pieces of first historical voice data is determined, wherein a first score of each piece of first target voice data of the at least one piece of first target voice data is greater than or equal to a preset threshold; a current threshold condition is acquired, wherein the current threshold condition is used for a condition whether the electronic deviceresponds to current voice data for operation; and the current threshold condition is adjusted based on the quantity of the at least one piece of first target voice data. The invention furthermore provides a data processing device applied to the electronic device, a data processing system and a computer readable storage medium.

The invention discloses an end-to-end voice awakening method and device and computer equipment. The method comprises the following steps: receiving voice to be detected; sequentially extracting acoustic features of each voice frame; inputting the extracted acoustic features into an acoustic model to obtain the probability of each target pronunciation unit in each voice frame output by the acousticmodel; calculating the cumulative probability of each target pronunciation unit frame by frame, and determining the starting position and the ending position of each target pronunciation unit; calculating the length and average probability of each target pronunciation unit according to the cumulative probability, the starting position and the ending position; and performing awakening operation according to the length and / or average probability of each target pronunciation unit. According to the invention, the false awakening rate can be reduced, and the awakening effect is improved.

The disclosure provides a wake-up word evaluation method and device, a storage medium, and electronic equipment. The method includes: acquiring a word to be evaluated input by a user; extracting evaluation features of the word to be evaluated, and the evaluation feature is used to represent the distinction of the word to be evaluated at an acoustic level and / or a semantic level; The evaluation feature of the evaluation word is used as an input, and after being processed by the pre-built wake-up word evaluation model, it is determined whether the word to be evaluated is suitable as a wake-up word. Such a solution helps to improve the accuracy of the wake-up word evaluation result, thereby improving the wake-up effect of the wake-up word set by the user.

The awakening method comprises the steps: judging whether the distance between a microphone array and a sound source point is greater than a preset distance or not; if yes, calculating calculation results of all preset wave beam directions in a microphone array on the current voice signal in parallel through a wave beam algorithm, wherein the microphone array is composed of a plurality of microphones evenly distributed on the same circular ring, and all the preset wave beam directions are all directions equally dividing the circumferential angle of the circular ring, and all the preset wave beam directions at least comprise four directions which are perpendicular to each other; according to calculation results of the current voice signals corresponding to the preset beam directions, calculating awakening probability values corresponding to the preset beam directions in parallel through an awakening algorithm; judging whether the wake-up probability value corresponding to each preset beam direction is greater than a first wake-up threshold value or not; if so, awakening according to the specified beam direction with the awakening probability greater than the first awakening threshold. Therefore, various defects caused by serial processing in the existing sound source positioning are avoided.

The invention provides a wake-up method, device and equipment for intelligent equipment. The method comprises the following steps: acquiring an audio data sequence; inputting the audio data sequence into a wake-up model, performing feature extraction on the audio data sequence by using the wake-up model to obtain a first sequence feature, and predicting whether a wake-up statement sequence is identified and outputting a corresponding sequence tag according to a second sequence feature and the first sequence feature of a wake-up statement sample used for waking up the intelligent device; and when it is determined that the sequence tag of the identified wake-up statement sequence is output, waking up the intelligent device. According to the invention, resources occupied by the wake-up function of the intelligent device can be reduced, the dependence of the wake-up function on alignment processing is reduced, and the wake-up effect is improved.

The awakening method comprises the steps: judging whether the distance between a microphone array and a sound source point is greater than a preset distance or not; if yes, calculating calculation results of all preset wave beam directions in a microphone array on the current voice signal in parallel through a wave beam algorithm, wherein the microphone array is composed of a plurality of microphones evenly distributed on the same circular ring, and all the preset wave beam directions are all directions equally dividing the circumferential angle of the circular ring, and all the preset wave beam directions at least comprise four directions which are perpendicular to each other; according to calculation results of the current voice signals corresponding to the preset beam directions, calculating awakening probability values corresponding to the preset beam directions in parallel through an awakening algorithm; judging whether the wake-up probability value corresponding to each preset beam direction is greater than a first wake-up threshold value or not; if so, awakening according to the specified beam direction with the awakening probability greater than the first awakening threshold. Therefore, various defects caused by serial processing in the existing sound source positioning are avoided.

The present invention relates to a method and device for improving wake-up performance. The method includes: obtaining a general reference model, wherein the general reference model is obtained by training a neural network with marked speech; obtaining wake-up speech of users of different age groups data; for the users of each age group, a custom model is generated through the general benchmark model and the wake-up voice data of the user; the custom model of the user of each age group is combined to generate a custom model for all age groups, to Wake up with the all-ages custom model. By customizing different neural network models for people of different age groups, the ability to distinguish models of different age groups can be improved in a targeted manner, and the wake-up performance of people of different age groups can be improved; Reusing most of the neural network calculations has a limited increase in computing resource consumption.

The invention provides a data processing method which comprises the following steps: obtaining current voice data; processing the current voice data according to historical voice data to obtain a current decibel value of the current voice data, wherein the historical voice data comprise at least one voice sub datum; determining whether or not the historical voice data comprises at least one targetvoice sub datum based on at least one of the current decibel value and the historical voice data, wherein the target voice sub datum comprises a voice datum for controlling failure of the electronicequipment; and responding to at least one target voice sub datum included in the historical voice data, storing at least one target voice sum datum as training for controlling sample data of a voice model for controlling the electronic equipment. The invention further provides a data processing device, a data processing system and a computer readable storage medium.

The invention provides an intelligent alarm clock and relates to the field of alarm clocks. The intelligent alarm clock comprises an alarm clock box, each of four corners of the bottom of the alarm clock box are fixedly connected with one end of a damper, the top end of the inside of each damper is fixedly connected with one end of a spring, the other end of each spring is fixedly connected with the top of a pressing block, the bottom of each pressing block is fixedly connected with the top end of a pressing rod, and the bottom of each pressing rod penetrates the corresponding damper and extends to space below the damper to be fixedly connected with the top of the base. Through the dampers at the four corners of the bottom of the alarm clock box, a motor positioned at the top of the base,a cam which is arranged, a cam seat positioned at the bottom of the alarm clock box, a supporting seat which is arranged and a sliding groove which is formed, when alarming time is up, the cam is driven to rotate to enable the alarm clock box to vibrate up and down along the sliding groove under action of the springs inside the dampers, so that alarming effect is improved, and people are enabled to be less prone to missing the alarming time.

The invention discloses a terminal equipment awakening method and device, a storage medium and an electronic device. The method comprises the steps: obtaining to-be-recognized audio data; in each wake-up model of at least two wake-up models configured in the terminal equipment, carrying out wake-up identification based on audio features under different dimensions extracted from the audio data, and acquiring audio identification results corresponding to the wake-up models, wherein each wake-up model is used for extracting audio features under one dimension; and under the condition that the audio identification results meet the wake-up condition, adjusting the terminal equipment to be in a wake-up state. By adopting the technical scheme, the problem of poor wake-up performance of the terminal equipment in the prior art is solved.

The invention provides a quilt with a wake-up function. The quilt comprises a cover, an inflation component and a first control component, the cover is provided with a non-breathable layer, a containing cavity is defined in the non-breathable layer and provided with an air tap, the inflation component is communicated with the air tap and used for inflating the containing cavity to bulge the cover,cold air enters the cover from the periphery of the cover and wakes up a user, and the first control component is connected with the inflation component to control opening or closing of the inflationcomponent. According to the quilt with the wake-up function, air in the containing cavity is exhausted by the inflation component, comfort of the user is reduced, the user is woken up in a tender mode, and the quilt is simple to use and convenient to operate.