CN110415724A - Transmission method, device, system and the computer readable storage medium of alert data - Google Patents
Transmission method, device, system and the computer readable storage medium of alert data Download PDFInfo
- Publication number
- CN110415724A CN110415724A CN201910729559.0A CN201910729559A CN110415724A CN 110415724 A CN110415724 A CN 110415724A CN 201910729559 A CN201910729559 A CN 201910729559A CN 110415724 A CN110415724 A CN 110415724A
- Authority
- CN
- China
- Prior art keywords
- alert data
- voice signal
- machine learning
- monitoring device
- learning algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
- G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
Abstract
The invention discloses a kind of transmission methods of alert data, listen to terminal applied to alarm, method includes the following steps: the voice signal of acquisition current environment；Feature extraction is carried out to voice signal, obtains feature extraction result；Test identification is carried out to feature extraction result using machine learning algorithm model, obtains test recognition result；When determining that voice signal includes medical monitoring device alert data according to test recognition result, medical monitoring device alert data is sent to user client.Using technical solution provided by the embodiment of the present invention, realize grasping and responding in time to Medical Devices alarm sound, medical care resource is significantly saved, the probability that privacy acoustic information is ravesdropping present in voice signal is significantly reduced, reduces data transmission cost.The invention also discloses a kind of transmitting device of alert data, equipment and storage mediums, have relevant art effect.
Description
Technical field
The present invention relates to sound processing techniques fields, more particularly to the transmission method, device, equipment of a kind of alert data
And computer readable storage medium.
Background technique
Hospital's scene carries out in surgical procedure, various Medical Devices booting work is had, to mention for surgical procedure or monitoring
For convenient, the electric knife on such as ventilator, heart beat monitor, infusion pump and operating table being commonly used.One operation is implemented
Process, which is often divided the work by the medical staff of different role, to be completed, and the Medical Devices that different role is used and paid close attention to are not also identical.In
Under one high-strung operating environment, and due to the difference of professional division, so that all kinds of medical care roles need entirely in surgical procedure
Participation that journey is seamless, and the moment pays close attention to each monitoring of tools situation, in order to avoid miss possible medical safety alarm.In fact, in one
Major surgery often continues the even longer time several hours, and having a meal, drink water, visiting the restroom etc. for period medical staff is necessary
The activity of solution be bound to cause supervision on temporary vacuum, especially to needs pay close attention to Medical Devices alarm sound and
When grasp and response lack, and if merely by manpower arrange an order according to class and grade maintain certainly will will increase consumption anxiety medical care resource.
In conclusion grasping and respond and medical care in time for Medical Devices alarm sound can not be accomplished by how efficiently solving
The problem of consumption of resource is balanced is current those skilled in the art's urgent problem.
Summary of the invention
The object of the present invention is to provide a kind of transmission methods of alert data, and the method achieve to Medical Devices alarm
Grasping and respond in time for sound, significantly saves medical care resource, significantly reduces privacy sound present in voice signal
The probability that information is ravesdropping reduces data transmission cost；It is a further object of the present invention to provide a kind of transmission of alert data
Device, equipment and computer readable storage medium.
In order to solve the above technical problems, the invention provides the following technical scheme:
A kind of transmission method of alert data, listens to terminal applied to alarm, comprising:
Acquire the voice signal of current environment；
Feature extraction is carried out to the voice signal, obtains feature extraction result；
Test identification is carried out to the feature extraction result using machine learning algorithm model, obtains test recognition result；
When determining that the voice signal includes medical monitoring device alert data according to the test recognition result, by institute
It states medical monitoring device alert data and is sent to user client.
In a kind of specific embodiment of the invention, the medical monitoring device alert data is sent to user client
End, comprising:
The medical monitoring device alert data is sent to cloud server, to utilize the cloud server described
Medical monitoring device alert data is transmitted to the user client.
In a kind of specific embodiment of the invention, further includes:
Receive model parameter after the adjustment that the cloud server returns；
Corresponding model parameter in the machine learning algorithm model is updated to model parameter after the adjustment.
In a kind of specific embodiment of the invention, further includes:
Receive the target machine learning algorithm model that the cloud server returns；
Former machine learning algorithm model is replaced with into the target machine learning algorithm model；
Wherein, the target machine learning algorithm model is to carry out model training to the medical monitoring device alert data
The test effect of output model up to standard.
A kind of transmitting device of alert data, listens to terminal applied to alarm, comprising:
Signal acquisition module, for acquiring the voice signal of current environment；
Characteristic extracting module obtains feature extraction result for carrying out feature extraction to the voice signal；
Identification module is tested, carries out test knowledge for the feature extraction result to be input to machine learning algorithm model
Not, test recognition result is obtained；
Alert data sending module determines that the voice signal includes medical treatment prison according to the test recognition result for working as
When protecting equipment alarm data, the medical monitoring device alert data is sent to user client.
A kind of Transmission system of alert data, comprising:
Terminal is listened in alarm, for acquiring the voice signal of current environment；Feature extraction is carried out to the voice signal, is obtained
To feature extraction result；Test identification is carried out to the feature extraction result using machine learning algorithm model, test is obtained and knows
Other result；
User client listens to terminal for receiving the alarm and determines sound letter according to the test recognition result
Number include medical monitoring device alert data when, the medical monitoring device alert data of transmission.
In a kind of specific embodiment of the invention, further includes:
Cloud server listens to terminal for receiving the alarm and determines sound letter according to the test recognition result
Number include medical monitoring device alert data when, the medical monitoring device alert data of transmission；
The user client, specifically for receiving the medical monitoring device alarm number of the cloud server forwarding
According to.
In a kind of specific embodiment of the invention, the cloud server is also used to generate model parameter after adjustment,
Model parameter after the adjustment is returned into the alarm and listens to terminal；
Terminal is listened in the alarm, is also used to the corresponding model parameter in the machine learning algorithm model being updated to institute
State model parameter after adjusting.
In a kind of specific embodiment of the invention, the cloud server is also used to obtain to the medical monitoring
The test effect of equipment alarm data progress model training output target machine learning algorithm model up to standard；By the target machine
Device learning algorithm model returns to the alarm and listens to terminal；
Terminal is listened in the alarm, is also used to former machine learning algorithm model replacing with the target machine learning algorithm
Model.
A kind of computer readable storage medium is stored with computer program on the computer readable storage medium, described
The step of transmission method of alert data as previously described is realized when computer program is executed by processor.
Using method provided by the embodiment of the present invention, terminal is listened to applied to alarm, acquires the sound letter of current environment
Number；Feature extraction is carried out to voice signal, obtains feature extraction result；Using machine learning algorithm model to feature extraction result
Test identification is carried out, test recognition result is obtained；When according to test recognition result determine voice signal include medical monitoring device
When alert data, medical monitoring device alert data is sent to user client.Machine is disposed in terminal by listening in alarm
Device learning algorithm model can in time be handled collected voice signal, when utilization machine learning algorithm model measurement
When recognizing voice signal comprising medical monitoring device alert data, medical monitoring device alert data is sent to user client
End.Grasping and responding in time to Medical Devices alarm sound is realized, medical care resource is significantly saved, by believing sound
Number processing after medical monitoring device alert data is only sent to user client, significantly reduce present in voice signal
The probability that privacy acoustic information is ravesdropping, reduces volume of transmitted data, to reduce data transmission cost.
Correspondingly, the embodiment of the invention also provides alert datas corresponding with the transmission method of above-mentioned alert data
Transmitting device, equipment and computer readable storage medium, have above-mentioned technique effect, and details are not described herein.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of implementation flow chart of the transmission method of alert data in the embodiment of the present invention；
Fig. 2 is another implementation flow chart of the transmission method of alert data in the embodiment of the present invention；
Fig. 3 is another implementation flow chart of the transmission method of alert data in the embodiment of the present invention；
Fig. 4 is a kind of structural block diagram of the transmitting device of alert data in the embodiment of the present invention；
Fig. 5 is a kind of structural block diagram of the Transmission system of alert data in the embodiment of the present invention；
Fig. 6 is another structural block diagram of the Transmission system of alert data in the embodiment of the present invention；
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
Embodiment one:
Referring to Fig. 1, Fig. 1 is a kind of implementation flow chart of the transmission method of alert data in the embodiment of the present invention, is applied to
Terminal is listened in alarm, and this method may comprise steps of:
S101: the voice signal of current environment is acquired.
Alarm can be set in medical monitoring device in the environment and listen to terminal, listen to terminal using alarm and acquire currently
The voice signal of environment.The voice signal of current environment may include personnel's telephone voice signal, the alarm of medical monitoring device
All voice signals including voice signal etc..Terminal is listened in alarm can be specifically set as sound pick-up.
S102: feature extraction is carried out to voice signal, obtains feature extraction result.
After getting the voice signal of current environment, feature extraction pretreatment can be carried out to voice signal, obtained
Feature extraction result.Characteristic extraction procedure may include converting voice signal by continuous analog signal s by modulus (AD)
(t) discrete sampling obtains digital signal s (n).Sub-frame processing is carried out to digital signal s (n), obtains frame signal si(n), i is indicated
I-th frame, during sub-frame processing, consecutive frame has lap.It is assumed that the sampling period is ts, overlapping number of samples is Nc, frame signal
Length NF, NF> Nc, then the consecutive frame shift time period is calculated as (NF-Nc)*ts.If tail frame effectively data length is inadequate, mend
Zero processing.Frame power Spectral Estimation: frame signal carries out discrete fourier DFT transform first, and formula indicates such as:
Wherein, h (n) is window function, is substantially carried out smoothing processing, and N indicates that discrete Fourier transform operation point is long, generally
2 power is such as 512 points long.By discrete Fourier transform calculated result, and then corresponding power spectrum is sought, and only retain the first half
Value, has:
The filtering of Meier filter group: Meier filter group is made of one group of triangular filter, might as well assume filter quantity
For Nf, filter length and power spectrum data etc. are a length ofMeier filter bank coefficients calculating process is as follows:
1) mel-frequency is calculated: by sampling period tsKnow sample frequency fs=1/ts, and then voice signal maximum frequency can be obtained
fmax=fs/ 2, it is assumed that human ear is f to the minimum frequency of sound sensitivemin.It is natural according to the mankind to the sensitive relations of sound frequency
Frequency translation is mel-frequency expression:
Mel-frequency expression range can be calculated to obtain:
2) Meier filter centre frequency calculates: mel-frequency being carried out to be divided into NfPart, mel-frequency stepping-in amount indicates
Are as follows:
Meier filter group centre frequency are as follows:
fmel(fmin)+Δfmel,fmel(fmin)+2*Δfmel,...,fmel(fmax)-Δfmel；
3) Meier centre frequency is converted to natural frequency: conversion formula is
N is calculated accordinglyf+ 2 frequency points, head and the tail frequency point are respectively fmin, fmax, intermediate frequency point is by last in step 2)
First formula that one formula substitutes into step 3) is calculated, and might as well assume that center frequency point is expressed as f (m) sequence, m=0,
1,...,Nf+ 1, then:
4) calculate Meier filter bank coefficients: calculation formula is expressed as follows, wherein m=1 ..., Nf, k=1 ..., Nf。
Calculate Meier filter group output: ai(m) indicate that the i-th frame data pass through output sequence after Meier filter group, m=
1,...,Nf。
MFCC characteristics extraction: to ai(m) logarithm operation is taken, and carries out discrete cosine transform, takes previous half data conduct
MFCC characteristic quantity indicates are as follows:
Wherein, floor () indicates rounding operation.
Interframe feature extraction: to improve recognition effect, consider frame-to-frame correlation, carry out feature expansion, supplement interframe feature
Are as follows:
Wherein, M indicates the frame number of backtracking.
Composite character Xi: Xi=[bi,ci], characteristic is defeated as data important in embedded machine learning module
Enter parameter.
S103: carrying out test identification to feature extraction result using machine learning algorithm model, obtains test recognition result.
Machine learning algorithm model can be stored in advance in embedded machine learning module, obtain feature extraction result it
Afterwards, it can use machine learning algorithm model and test identification carried out to feature extraction result, obtain test recognition result.In undertaking
Citing is stated, the process that machine learning algorithm model carries out test identification to feature extraction result may include: machine learning algorithm
Model can indicate that model parameter is indicated with H with F, carry out model reasoning calculating after composite character input, export calculated result
Yi, mathematical expression are as follows:
Yi=F (Xi,H)；
In exemplary application, embedded machine learning module can select the neural calculation rod NCS such as Intel, tall and handsome to reach
Jetson TX series module etc..Wherein, NCS supports convolutional neural networks deep learning frame (Caffe, Convolutional
Architecture for Fast Feature Embedding), Jetson also supports configuration tensorflow deep learning frame
Frame supports more algorithm models to select.For medical monitoring device work alarm sound monitoring, after locally carrying out model training,
Algorithm model selection and corresponding algorithm model parameter can be preset in embedded machine learning module, after powering on automatic load or
Implement to configure by cloud in pilot process.
S104:, will be medical when determining that voice signal includes medical monitoring device alert data according to test recognition result
Custodial care facility alert data is sent to user client.
Obtain test recognition result after, when according to test recognition result determine voice signal include medical monitoring device
When alert data, medical monitoring device alert data is sent to user client.Terminal is listened in alarm can be by wired side
Formula and user client, which are established, to be communicated to connect, and can also establish communication connection wirelessly, the embodiment of the present invention to this not
It limits.
User client is mainly mobile client software, may include APP, small routine, and the softwares such as wechat public platform are in
Existing form, user client system is mainly mounted in a software form in mobile phone or other mobile devices (such as plate), before UI
End is presented to the user observation and interactive controlling button uses, and covers user management, listens to the subfunctions such as alarm setting, training mark
Module.Wherein user management can be related to the operation such as user's registration, cancellation, account management；It listens to alarm setting and is primarily upon report
The operation such as setting, change, cancellation of police's formula；Training labeling module is by mobile device microphone authority acquiring, with software
Instruction mode opens mobile device data acquisition service, completes analog-to-digital conversion in mobile device hardware, to obtain trained number
According to and can implement to mark manually, primary data is uploaded to cloud server system finally by user client Transmission system.
User client Transmission system establishes data and instruction communication interaction by network and cloud server system.User client is handed over
The unified interactive controlling that mutual control system implements client software is coordinated.
Alert data transmission plan provided by the embodiment of the present invention significantly enhances the convenience of remote operation, is
Medical staff provides reliable and secure alert detecting under scene and entrusts in of short duration leave, medical staff can more trust have a meal,
It the activities such as drinks water, visit the restroom, having reached grasping and respond and the consumption of medical care resource in time for Medical Devices alarm sound
Balance, promotes medical and nursing work efficiency, while enhancing medical safety.By listening to terminal in alarm, i.e. data source header is to adopting
After the voice signal collected is handled, so that the voice data uploaded is only medical monitoring device alert data, privacy content
Automated intelligent shielding significantly reduces (the dialogue letter between such as medical staff of privacy acoustic information present in voice signal
Breath) probability that is ravesdropping, and significantly reduce volume of transmitted data, reduce flow and use, thus reduce data transmission at
This, improves the timeliness of alert data transmission.
Using method provided by the embodiment of the present invention, terminal is listened to applied to alarm, acquires the sound letter of current environment
Number；Feature extraction is carried out to voice signal, obtains feature extraction result；Using machine learning algorithm model to feature extraction result
Test identification is carried out, test recognition result is obtained；When according to test recognition result determine voice signal include medical monitoring device
When alert data, medical monitoring device alert data is sent to user client.Machine is disposed in terminal by listening in alarm
Device learning algorithm model can in time be handled collected voice signal, when utilization machine learning algorithm model measurement
When recognizing voice signal comprising medical monitoring device alert data, medical monitoring device alert data is sent to user client
End.Grasping and responding in time to Medical Devices alarm sound is realized, medical care resource is significantly saved, by believing sound
Number processing after medical monitoring device alert data is only sent to user client, significantly reduce present in voice signal
The probability that privacy acoustic information is ravesdropping, reduces volume of transmitted data, to reduce data transmission cost.
It should be noted that based on the above embodiment one, the embodiment of the invention also provides be correspondingly improved scheme.Rear
Involved in continuous embodiment with can mutually be referred between same steps or corresponding steps in above-described embodiment one, corresponding beneficial effect
Can also be cross-referenced, it is no longer repeated one by one in improvement embodiment below.
Embodiment two:
Referring to fig. 2, Fig. 2 is another implementation flow chart of the transmission method of alert data in the embodiment of the present invention, application
Terminal is listened in alarm, this method may comprise steps of:
S201: the voice signal of current environment is acquired.
S202: feature extraction is carried out to voice signal, obtains feature extraction result.
S203: carrying out test identification to feature extraction result using machine learning algorithm model, obtains test recognition result.
S204:, will be medical when determining that voice signal includes medical monitoring device alert data according to test recognition result
Custodial care facility alert data is sent to cloud server, to be transmitted to medical monitoring device alert data using cloud server
User client.
When determining that voice signal includes medical monitoring device alert data according to test recognition result, medical treatment can be supervised
Shield equipment alarm data are sent to cloud server, and medical monitoring device alert data is transmitted to user visitor again by cloud server
Family end.Cloud server can be mainly by display system, storage system, control system, machine learning system and Transmission system group
At display system can be made of the cloud front end UI, and main addition and deletion convenient for user client carries out cloud management.
Storage system can be made of Database Systems, be stored to medical monitoring device alert data, can in machine learning system
(it is equal with the embedded machine learning module that terminal is listened in alarm and supports, to ensure that model is consistent disposes machine learning frame
Property), it is substantially carried out machine learning model training and test, Transmission system is responsible for listening to terminal system, user client with alarm
System carries out data and instruction communication and delivery, and the control that cloud control system is responsible for implementing between each module in cloud is coordinated.
The networking mode of terminal and cloud server is listened in alarm, and there are many modes, and not limiting use according to application conditions has
Gauze network or wireless network mode, such as WIFI and general mobile communications network, the connection of user client and cloud server
Net mode mainly by user client online condition setting, mainly can realize networking work by mobile communications network or WIFI
Make.
S205: model parameter after the adjustment that cloud server returns is received.
Terminal is listened in alarm can receive model parameter after the adjustment that cloud server returns.Accept the act in step S104
Example, after training is implemented to mark and be uploaded to cloud server by user client with data, cloud server can be extracted
Trained voice data and training annotation resultsSpeech feature extraction is carried out, composite character is generatedIts voice
Characteristic extraction procedure is referred to the speech feature extraction process that terminal is listened in alarm.According to composite characterIt is marked with corresponding
As a resultModel best identified parameter { H is calculated under certain algorithm constraint conditionC, that is, model parameter after adjusting is full
Sufficient mathematic(al) representation:
S206: the corresponding model parameter in machine learning algorithm model is updated to model parameter after adjusting.
After being adjusted after model parameter, the corresponding model parameter in machine learning algorithm model can be updated to
Model parameter after adjustment supervises medical treatment with optimal identification parameter so that the machine learning algorithm model of terminal is listened in alarm
Shield equipment alarm data carry out test identification.
Embodiment three:
Referring to Fig. 3, Fig. 3 is another implementation flow chart of the transmission method of alert data in the embodiment of the present invention, application
Terminal is listened in alarm, this method may comprise steps of:
S301: the voice signal of current environment is acquired.
S302: feature extraction is carried out to voice signal, obtains feature extraction result.
S303: carrying out test identification to feature extraction result using machine learning algorithm model, obtains test recognition result.
S304:, will be medical when determining that voice signal includes medical monitoring device alert data according to test recognition result
Custodial care facility alert data is sent to cloud server, to be transmitted to medical monitoring device alert data using cloud server
User client.
S305: the target machine learning algorithm model that cloud server returns is received.
Machine learning algorithm model can be preset medical monitoring device alert data is carried out to identify that need are to be achieved
Test effect standard.If test effect is not up to standard, cloud after the trained ginseng with tune of the machine learning algorithm model of current selected
Server can replace other machine learning algorithm model re -trainings, obtain carrying out mould to medical monitoring device alert data
The test effect of type training output target machine learning algorithm model up to standard, and target machine learning algorithm model is sent to
Terminal is listened in alarm, and alarm listens to terminal and receives the target machine learning algorithm model that cloud server returns.
S306: former machine learning algorithm model is replaced with into target machine learning algorithm model.
Wherein, target machine learning algorithm model is the survey that model training output is carried out to medical monitoring device alert data
Try effect model up to standard.
After receiving target machine learning algorithm model, former machine learning algorithm model can be replaced with into target machine
Device learning algorithm model.
Corresponding to above method embodiment, the embodiment of the invention also provides a kind of transmitting device of alert data, under
The transmitting device of the alert data of text description can correspond to each other reference with the transmission method of above-described alert data.
Referring to fig. 4, Fig. 4 is a kind of structural block diagram of the transmitting device of alert data in the embodiment of the present invention, is applied to report
Police listens to terminal, the apparatus may include:
Signal acquisition module 41, for acquiring the voice signal of current environment；
Characteristic extracting module 42 obtains feature extraction result for carrying out feature extraction to voice signal；
Identification module 43 is tested, carries out test identification for feature extraction result to be input to machine learning algorithm model,
Obtain test recognition result；
Alert data sending module 44 determines that voice signal includes medical monitoring device according to test recognition result for working as
When alert data, medical monitoring device alert data is sent to user client.
Using device provided by the embodiment of the present invention, terminal is listened to applied to alarm, acquires the sound letter of current environment
Number；Feature extraction is carried out to voice signal, obtains feature extraction result；Using machine learning algorithm model to feature extraction result
Test identification is carried out, test recognition result is obtained；When according to test recognition result determine voice signal include medical monitoring device
When alert data, medical monitoring device alert data is sent to user client.Machine is disposed in terminal by listening in alarm
Device learning algorithm model can in time be handled collected voice signal, when utilization machine learning algorithm model measurement
When recognizing voice signal comprising medical monitoring device alert data, medical monitoring device alert data is sent to user client
End.Grasping and responding in time to Medical Devices alarm sound is realized, medical care resource is significantly saved, by believing sound
Number processing after medical monitoring device alert data is only sent to user client, significantly reduce present in voice signal
The probability that privacy acoustic information is ravesdropping, reduces volume of transmitted data, to reduce data transmission cost.
In a kind of specific embodiment of the invention, alert data sending module 44 is specially by medical monitoring device report
Alert data are sent to cloud server, medical monitoring device alert data is transmitted to user client using cloud server
Module.
In a kind of specific embodiment of the invention, which can also include:
Parameter receiving module, model parameter after the adjustment for receiving cloud server return；
Parameter updating module is joined for model after the corresponding model parameter in machine learning algorithm model is updated to adjustment
Number.
In a kind of specific embodiment of the invention, which can also include:
Model receiving module, for receiving the target machine learning algorithm model of cloud server return；
Model replacement module, for former machine learning algorithm model to be replaced with target machine learning algorithm model；
Wherein, target machine learning algorithm model is the survey that model training output is carried out to medical monitoring device alert data
Try effect model up to standard.
Corresponding to above method embodiment, the embodiment of the invention also provides a kind of Transmission system of alert data, under
The Transmission system of the alert data of text description can correspond to each other reference with the transmission method of above-described alert data.
Referring to Fig. 5, Fig. 5 is a kind of structural block diagram of the Transmission system of alert data in the embodiment of the present invention, which can
To include:
Terminal 1 is listened in alarm, for acquiring the voice signal of current environment；Feature extraction is carried out to voice signal, is obtained
Feature extraction result；Test identification is carried out to feature extraction result using machine learning algorithm model, obtains test recognition result；
User client 2 listens to terminal 1 for reception alarm and determines that voice signal includes medical treatment according to test recognition result
When custodial care facility alert data, the medical monitoring device alert data of transmission.
Using system provided by the embodiment of the present invention, terminal is listened to applied to alarm, acquires the sound letter of current environment
Number；Feature extraction is carried out to voice signal, obtains feature extraction result；Using machine learning algorithm model to feature extraction result
Test identification is carried out, test recognition result is obtained；When according to test recognition result determine voice signal include medical monitoring device
When alert data, medical monitoring device alert data is sent to user client.Machine is disposed in terminal by listening in alarm
Device learning algorithm model can in time be handled collected voice signal, when utilization machine learning algorithm model measurement
When recognizing voice signal comprising medical monitoring device alert data, medical monitoring device alert data is sent to user client
End.Grasping and responding in time to Medical Devices alarm sound is realized, medical care resource is significantly saved, by believing sound
Number processing after medical monitoring device alert data is only sent to user client, significantly reduce present in voice signal
The probability that privacy acoustic information is ravesdropping, reduces volume of transmitted data, to reduce data transmission cost.
In a kind of specific embodiment of the invention, which can also include:
Cloud server 3 listens to terminal 1 for reception alarm and determines that voice signal includes medical treatment according to test recognition result
When custodial care facility alert data, the medical monitoring device alert data of transmission；
User client 2, the medical monitoring device alert data forwarded specifically for receiving cloud server 3.
In a kind of specific embodiment of the invention, cloud server 3 is also used to generate model parameter after adjustment, will
Model parameter returns to alarm and listens to terminal 1 after adjustment；
Terminal 1 is listened in alarm, is also used to for the corresponding model parameter in machine learning algorithm model being updated to adjustment rear mold
Shape parameter.
In a kind of specific embodiment of the invention, cloud server 3 is also used to obtain and alarm medical monitoring device
The test effect of data progress model training output target machine learning algorithm model up to standard；By target machine learning algorithm mould
Type returns to alarm and listens to terminal 1；
Terminal 1 is listened in alarm, is also used to former machine learning algorithm model replacing with target machine learning algorithm model.
It is computer-readable the present invention also provides a kind of computer readable storage medium corresponding to above method embodiment
It is stored with computer program on storage medium, can realize following steps when computer program is executed by processor:
Acquire the voice signal of current environment；Feature extraction is carried out to voice signal, obtains feature extraction result；Utilize machine
Device learning algorithm model carries out test identification to feature extraction result, obtains test recognition result；When according to test recognition result
When determining that voice signal includes medical monitoring device alert data, medical monitoring device alert data is sent to user client
End.
The computer readable storage medium may include: USB flash disk, mobile hard disk, read-only memory (Read-Only
Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. is various to deposit
Store up the medium of program code.
Above method embodiment is please referred to for the introduction of computer readable storage medium provided by the invention, the present invention exists
This is not repeated them here.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
It sets, for equipment and computer readable storage medium, since it is corresponded to the methods disclosed in the examples, so the comparison of description
Simply, reference may be made to the description of the method.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand technical solution of the present invention and its core concept.It should be pointed out that for the common of the art
, without departing from the principle of the present invention, can be with several improvements and modifications are made to the present invention for technical staff, these
Improvement and modification are also fallen within the protection scope of the claims of the present invention.
Claims (10)
1. a kind of transmission method of alert data, which is characterized in that listen to terminal applied to alarm, comprising:
Acquire the voice signal of current environment；
Feature extraction is carried out to the voice signal, obtains feature extraction result；
Test identification is carried out to the feature extraction result using machine learning algorithm model, obtains test recognition result；
When determining that the voice signal includes medical monitoring device alert data according to the test recognition result, by the doctor
It treats custodial care facility alert data and is sent to user client.
2. the transmission method of alert data according to claim 1, which is characterized in that the medical monitoring device is alarmed
Data are sent to user client, comprising:
The medical monitoring device alert data is sent to cloud server, to utilize the cloud server by the medical treatment
Custodial care facility alert data is transmitted to the user client.
3. the transmission method of alert data according to claim 2, which is characterized in that further include:
Receive model parameter after the adjustment that the cloud server returns；
Corresponding model parameter in the machine learning algorithm model is updated to model parameter after the adjustment.
4. the transmission method of alert data according to claim 2 or 3, which is characterized in that further include:
Receive the target machine learning algorithm model that the cloud server returns；
Former machine learning algorithm model is replaced with into the target machine learning algorithm model；
Wherein, the target machine learning algorithm model is to carry out model training output to the medical monitoring device alert data
Test effect model up to standard.
5. a kind of transmitting device of alert data, which is characterized in that listen to terminal applied to alarm, comprising:
Signal acquisition module, for acquiring the voice signal of current environment；
Characteristic extracting module obtains feature extraction result for carrying out feature extraction to the voice signal；
Identification module is tested, test identification is carried out for the feature extraction result to be input to machine learning algorithm model, obtains
To test recognition result；
Alert data sending module determines that the voice signal is set comprising medical monitoring according to the test recognition result for working as
When standby alert data, the medical monitoring device alert data is sent to user client.
6. a kind of Transmission system of alert data characterized by comprising
Terminal is listened in alarm, for acquiring the voice signal of current environment；Feature extraction is carried out to the voice signal, obtains spy
Sign extracts result；Test identification is carried out to the feature extraction result using machine learning algorithm model, obtains test identification knot
Fruit；
User client listens to terminal and determines the voice signal packet according to the test recognition result for receiving the alarm
When alert data containing medical monitoring device, the medical monitoring device alert data of transmission.
7. the Transmission system of alert data according to claim 6, which is characterized in that further include:
Cloud server listens to terminal and determines the voice signal packet according to the test recognition result for receiving the alarm
When alert data containing medical monitoring device, the medical monitoring device alert data of transmission；
The user client, specifically for receiving the medical monitoring device alert data of the cloud server forwarding.
8. the Transmission system of alert data according to claim 7, which is characterized in that the cloud server is also used to
Model parameter after adjusting is generated, model parameter after the adjustment is returned into the alarm and listens to terminal；
Terminal is listened in the alarm, is also used to the corresponding model parameter in the machine learning algorithm model being updated to the tune
Model parameter after whole.
9. the Transmission system of alert data according to claim 7 or 8, which is characterized in that the cloud server is also used
Learn in the target machine for obtaining up to standard to the test effect of medical monitoring device alert data progress model training output
Algorithm model；The target machine learning algorithm model is returned into the alarm and listens to terminal；
Terminal is listened in the alarm, is also used to former machine learning algorithm model replacing with the target machine learning algorithm mould
Type.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program realizes the transmission side of the alert data as described in any one of Claims 1-4 when the computer program is executed by processor
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910729559.0A CN110415724A (en) | 2019-08-08 | 2019-08-08 | Transmission method, device, system and the computer readable storage medium of alert data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910729559.0A CN110415724A (en) | 2019-08-08 | 2019-08-08 | Transmission method, device, system and the computer readable storage medium of alert data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110415724A true CN110415724A (en) | 2019-11-05 |
Family
ID=68366699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910729559.0A Pending CN110415724A (en) | 2019-08-08 | 2019-08-08 | Transmission method, device, system and the computer readable storage medium of alert data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110415724A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179527A (en) * | 2019-12-26 | 2020-05-19 | 数海信息技术有限公司 | Alarm method, device and system based on dynamic audio information and storage medium |
| CN111243623A (en) * | 2019-12-26 | 2020-06-05 | 数海信息技术有限公司 | Progressive audio alarm method, device and system |
| CN111259461A (en) * | 2020-01-20 | 2020-06-09 | 北京科芮智盈技术有限公司 | Anti-stealing-lighting method, terminal equipment and system |
| CN112908357A (en) * | 2021-01-19 | 2021-06-04 | 广州橙行智动汽车科技有限公司 | Instrument sound testing method and device, testing equipment and storage medium |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070183604A1 (en) * | 2006-02-09 | 2007-08-09 | St-Infonox | Response to anomalous acoustic environments |
| CN103875034A (en) * | 2011-10-17 | 2014-06-18 | 皇家飞利浦有限公司 | A medical feedback system based on sound analysis in a medical environment |
| CN107122807A (en) * | 2017-05-24 | 2017-09-01 | 努比亚技术有限公司 | A kind of family's monitoring method, service end and computer-readable recording medium |
| CN107133612A (en) * | 2017-06-06 | 2017-09-05 | 河海大学常州校区 | Based on image procossing and the intelligent ward of speech recognition technology and its operation method |
| CN107222865A (en) * | 2017-04-28 | 2017-09-29 | 北京大学 | The communication swindle real-time detection method and system recognized based on suspicious actions |
| CN107247945A (en) * | 2017-07-04 | 2017-10-13 | 刘艺晴 | A kind of ward sufferer monitor system and monitoring method based on Kinect device |
| CN107610702A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Terminal device standby wakeup method, apparatus and computer equipment |
| CN107707657A (en) * | 2017-09-30 | 2018-02-16 | 苏州涟漪信息科技有限公司 | Safety custody system based on multisensor |
| CN108010537A (en) * | 2017-12-30 | 2018-05-08 | 长沙迪普美医疗科技有限公司 | A kind of audible alarm processing method, device, system |
| CN108511086A (en) * | 2018-04-13 | 2018-09-07 | 广州红象医疗科技有限公司 | Medical monitoring method and system, computer storage media and equipment |
| CN109473113A (en) * | 2018-11-13 | 2019-03-15 | 北京物灵智能科技有限公司 | A kind of sound identification method and device |
| CN109658953A (en) * | 2019-01-12 | 2019-04-19 | 深圳先进技术研究院 | A kind of vagitus recognition methods, device and equipment |
| CN109803207A (en) * | 2017-11-17 | 2019-05-24 | 英特尔公司 | Mark to the audio signal in ambient sound and the guidance to the autonomous vehicles in response to the mark |
| CN208903379U (en) * | 2018-08-17 | 2019-05-24 | 深圳伊艾克斯信息技术有限公司 | A kind of anti-theft alarm system based on face and voice recognition |
| CN109863553A (en) * | 2016-04-07 | 2019-06-07 | M.S.T.医学外科技术有限公司 | The operation control system of voice activation |
| CN110097876A (en) * | 2018-01-30 | 2019-08-06 | 阿里巴巴集团控股有限公司 | Voice wakes up processing method and is waken up equipment |
-
2019
- 2019-08-08 CN CN201910729559.0A patent/CN110415724A/en active Pending
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070183604A1 (en) * | 2006-02-09 | 2007-08-09 | St-Infonox | Response to anomalous acoustic environments |
| CN103875034A (en) * | 2011-10-17 | 2014-06-18 | 皇家飞利浦有限公司 | A medical feedback system based on sound analysis in a medical environment |
| CN109863553A (en) * | 2016-04-07 | 2019-06-07 | M.S.T.医学外科技术有限公司 | The operation control system of voice activation |
| CN107222865A (en) * | 2017-04-28 | 2017-09-29 | 北京大学 | The communication swindle real-time detection method and system recognized based on suspicious actions |
| CN107122807A (en) * | 2017-05-24 | 2017-09-01 | 努比亚技术有限公司 | A kind of family's monitoring method, service end and computer-readable recording medium |
| CN107133612A (en) * | 2017-06-06 | 2017-09-05 | 河海大学常州校区 | Based on image procossing and the intelligent ward of speech recognition technology and its operation method |
| CN107247945A (en) * | 2017-07-04 | 2017-10-13 | 刘艺晴 | A kind of ward sufferer monitor system and monitoring method based on Kinect device |
| CN107610702A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Terminal device standby wakeup method, apparatus and computer equipment |
| CN107707657A (en) * | 2017-09-30 | 2018-02-16 | 苏州涟漪信息科技有限公司 | Safety custody system based on multisensor |
| CN109803207A (en) * | 2017-11-17 | 2019-05-24 | 英特尔公司 | Mark to the audio signal in ambient sound and the guidance to the autonomous vehicles in response to the mark |
| CN108010537A (en) * | 2017-12-30 | 2018-05-08 | 长沙迪普美医疗科技有限公司 | A kind of audible alarm processing method, device, system |
| CN110097876A (en) * | 2018-01-30 | 2019-08-06 | 阿里巴巴集团控股有限公司 | Voice wakes up processing method and is waken up equipment |
| CN108511086A (en) * | 2018-04-13 | 2018-09-07 | 广州红象医疗科技有限公司 | Medical monitoring method and system, computer storage media and equipment |
| CN208903379U (en) * | 2018-08-17 | 2019-05-24 | 深圳伊艾克斯信息技术有限公司 | A kind of anti-theft alarm system based on face and voice recognition |
| CN109473113A (en) * | 2018-11-13 | 2019-03-15 | 北京物灵智能科技有限公司 | A kind of sound identification method and device |
| CN109658953A (en) * | 2019-01-12 | 2019-04-19 | 深圳先进技术研究院 | A kind of vagitus recognition methods, device and equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179527A (en) * | 2019-12-26 | 2020-05-19 | 数海信息技术有限公司 | Alarm method, device and system based on dynamic audio information and storage medium |
| CN111243623A (en) * | 2019-12-26 | 2020-06-05 | 数海信息技术有限公司 | Progressive audio alarm method, device and system |
| CN111259461A (en) * | 2020-01-20 | 2020-06-09 | 北京科芮智盈技术有限公司 | Anti-stealing-lighting method, terminal equipment and system |
| CN112908357A (en) * | 2021-01-19 | 2021-06-04 | 广州橙行智动汽车科技有限公司 | Instrument sound testing method and device, testing equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110415724A (en) | Transmission method, device, system and the computer readable storage medium of alert data | |
| CN104574617B (en) | Building gate inhibition safety management system | |
| CN106709254B (en) | A kind of medical diagnosis robot system | |
| CN108335696A (en) | Voice awakening method and device | |
| CN110176226A (en) | A kind of speech recognition and speech recognition modeling training method and device | |
| CN106707788B (en) | A kind of intelligent home voice control identifying system and method | |
| CN111739553B (en) | Conference sound collection, conference record and conference record presentation method and device | |
| CN106921911B (en) | Voice acquisition method and device | |
| CN109447156A (en) | Method and apparatus for generating model | |
| CN106934750A (en) | A kind of gymnasium management system | |
| CN109065046A (en) | Method, apparatus, electronic equipment and the computer readable storage medium that voice wakes up | |
| CN104352248A (en) | Heart sound, lung sound and fetal heart sound acquisition and health management system and method | |
| CN110189756A (en) | It is a kind of for monitoring the method and system of live pig abnormal sound | |
| CN109145828A (en) | Method and apparatus for generating video classification detection model | |
| CN107835490A (en) | Floor position localization method, system, server and computer-readable recording medium | |
| CN110720946A (en) | Intelligent auscultation system based on deep learning | |
| CN104042208B (en) | Electrocardiogram monitor system | |
| CN109524013A (en) | A kind of method of speech processing, device, medium and smart machine | |
| CN104462786A (en) | Networked intelligent multi-parameter vital sign monitoring device and cloud platform system | |
| CN110211121A (en) | Method and apparatus for pushing model | |
| CN104754112A (en) | User information obtaining method and mobile terminal | |
| CN105407155B (en) | A kind of ward is registered method and system | |
| CN107077531A (en) | Stethoscope data processing method and device, electronic equipment and cloud server | |
| CN114117053A (en) | Disease classification model training method and device, storage medium and electronic device | |
| CN203763055U (en) | Health monitoring alarm and help system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |