CN110136725B - Sign monitoring system through voice input - Google Patents
Sign monitoring system through voice input Download PDFInfo
- Publication number
- CN110136725B CN110136725B CN201810131046.5A CN201810131046A CN110136725B CN 110136725 B CN110136725 B CN 110136725B CN 201810131046 A CN201810131046 A CN 201810131046A CN 110136725 B CN110136725 B CN 110136725B
- Authority
- CN
- China
- Prior art keywords
- voice
- library
- input
- module
- early warning
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G10L15/28—Constructional details of speech recognition systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
Abstract
The invention relates to a sign monitoring system through voice input, which comprises a voice input module, a central processing module and a voice output module, wherein the voice input module is connected with the central processing module; the method is characterized in that: the central processing module is integrated with a voice recognition module, a judgment module and an early warning module; the voice recognition module comprises a keyword matching part, a physical sign type selection part and a physical sign type configuration library; the judgment module comprises a numerical value integrity judgment part, a numerical value abnormality judgment part and a retest reminding part; the early warning module comprises a threshold judging part, an early warning level selecting part, a health care knowledge judging part and a health care knowledge base; the voice parameter input method is convenient for old people who are not familiar with operation methods and old people to carry out voice parameter input through voice and are one of important targets of family sign monitoring, meanwhile, input voice is matched and identified through priority and relevancy, the precision is high, and the result is accurate.
Description
Technical Field
The invention relates to the field of physical sign monitoring, in particular to a physical sign monitoring system through voice input.
Background
With the development of the internet and the internet of things technology, various physical sign monitoring technologies based on the internet and the internet of things technology emerge endlessly, but due to lack of unified standards, different products, products of different models of the same brand, and products of different versions of the same model all need to be manually input again, and the repeated workload is too large and unstable. Meanwhile, the physical sign parameters are input through the internet of things technologies such as Bluetooth and network, and certain operations are needed, so that great troubles are caused to the old who are one of important targets of family physical sign monitoring, unfamiliar with an operation method and old.
Disclosure of Invention
The technical problem to be solved by the invention is; in order to overcome the defects and shortcomings in the prior art, the invention provides the sign monitoring system through voice input, which is convenient for old people who are not familiar with an operation method and old to carry out voice parameter input through voice and are one of important targets of family sign monitoring, and simultaneously carries out matching identification on input voice through priority and relevancy, and has high precision and accurate result.
The technical scheme adopted by the invention for solving the technical problems is as follows: a sign monitoring system through voice input comprises a voice input module, a central processing module and a voice output module, wherein the voice input module is connected with the central processing module, and the central processing module is connected with the voice output module; the method is characterized in that: the central processing module is integrated with a voice recognition module, a judgment module and an early warning module;
the voice recognition module comprises a keyword matching part, a physical sign type selection part and a physical sign type configuration library;
the judgment module comprises a numerical value integrity judgment part, a numerical value abnormality judgment part and a retest reminding part;
the early warning module comprises a threshold judging part, an early warning level selecting part, a health care knowledge judging part and a health care knowledge base.
Furthermore, a matching unit, a keyword library, a local dialect library, a fuzzy voice library and a personal voice library are arranged in the keyword matching part; the matching unit searches relevant words of the input voice from the keyword library, the local dialect library, the fuzzy voice library and the personal voice library and finally forms matching voice.
Furthermore, a priority judging unit and a correlation judging unit are further arranged in the keyword matching part, the priority judging unit automatically forms a priority level of a matching unit for searching related vocabularies of input voice from the keyword library, the local dialect library, the fuzzy voice library and the personal voice library according to a preset result and a later-stage adjusting result, the correlation judging unit forms the correlation degree with the keyword library, the local dialect library, the fuzzy voice library and the personal voice library according to a result of finally forming the matched voice each time, and the priority judging unit is adjusted later according to the correlation degree.
Further, the sign types in the sign type selection part comprise blood pressure, heart rate, body temperature, height/weight and blood sugar information.
Furthermore, the physical sign type configuration library can be increased or decreased through an external device, and the external device comprises a mobile hard disk, a mobile phone, a watch, an ipad or a computer which can be connected to a network.
Further, a numerical integrity threshold and a voice example prompt are set in the numerical integrity judgment section, and the voice example prompt is issued when the voice input does not meet the numerical integrity threshold.
Furthermore, a numerical value abnormality determination unit is arranged in the numerical value abnormality degree determination part, and after the numerical value abnormality determination unit determines that the input voice is abnormal, information is sent to the retesting reminding part to remind the user of performing voice input again.
Further, the threshold determination unit performs threshold determination on the input voice based on a preset threshold; the early warning level selection part determines the early warning level according to the part of the input voice exceeding the threshold value, and the health care knowledge judgment part selects the corresponding health care knowledge according to the type of the input voice exceeding the threshold value.
Furthermore, the health care knowledge base is connected with the hospital background and can be synchronized with the hospital background to update the health care knowledge base.
The invention has the advantages that;
(1) the voice parameter input is conveniently carried out by the elderly who are not familiar with the operation method and are one of important targets of family sign monitoring through voice, so that troubles caused by the limitation of an input mode to the health sign monitoring of the elderly are avoided;
(2) meanwhile, the input voice is matched and identified through the priority and the relevancy, the precision is high, and the result is accurate.
Drawings
Fig. 1 is a schematic structural diagram of a sign monitoring system through voice input according to the present invention.
Detailed Description
The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution related to the present invention.
As shown in fig. 1, a sign monitoring system by voice input includes a voice input module, a central processing module and a voice output module, wherein the voice input module is connected with the central processing module, and the central processing module is connected with the voice output module; the method is characterized in that: the central processing module is integrated with a voice recognition module, a judgment module and an early warning module;
the voice recognition module comprises a keyword matching part, a physical sign type selection part and a physical sign type configuration library;
the judgment module comprises a numerical value integrity judgment part, a numerical value abnormality judgment part and a retest reminding part;
the early warning module comprises a threshold judging part, an early warning level selecting part, a health care knowledge judging part and a health care knowledge base.
Specifically, the keyword matching part is provided with a matching unit, a keyword library, a local dialect library, a fuzzy voice library and a personal voice library; the matching unit searches relevant words of the input voice from the keyword library, the local dialect library, the fuzzy voice library and the personal voice library and finally forms matching voice.
Specifically, a priority judging unit and a relevancy judging unit are further arranged in the keyword matching part, the priority judging unit automatically forms a priority level of a relevant vocabulary of the input voice searched by the matching unit from the keyword library, the local dialect library, the fuzzy voice library and the personal voice library according to a preset result and a later-stage adjusting result, the relevancy judging unit forms the relevancy with the keyword library, the local dialect library, the fuzzy voice library and the personal voice library according to a result of finally forming the matched voice every time, and the later-stage adjusting unit is adjusted according to the relevancy.
Specifically, the sign types in the sign type selection part include blood pressure, heart rate, body temperature, height/weight and blood sugar information.
Specifically, the physical sign type configuration library can be increased or decreased through an external device, and the external device comprises a mobile hard disk, a mobile phone, a watch, an ipad or a computer which can be connected to a network.
Specifically, the numerical integrity judgment unit is provided with a numerical integrity threshold value and a voice example prompt, and the voice example prompt is issued when the voice input does not meet the numerical integrity threshold value.
Specifically, a numerical value abnormality determination unit is arranged in the numerical value abnormality degree determination unit, and after the numerical value abnormality determination unit determines that the input voice is abnormal, information is sent to the retest reminding unit to remind the user of performing voice input again.
Specifically, the threshold determination unit performs threshold determination on the input speech sound based on a preset threshold; the early warning level selection part determines the early warning level according to the part of the input voice exceeding the threshold value, and the health care knowledge judgment part selects the corresponding health care knowledge according to the type of the input voice exceeding the threshold value.
Specifically, the health care knowledge base is connected with the hospital background and can be synchronized with the hospital background to update the health care knowledge base.
Specifically, the monitoring method of the voice-input physical sign monitoring system comprises the following steps:
(1) sending out voice prompt to guide the user to input voice;
(2) a user sends parameters representing own physical signs through a voice input module;
(3) the voice input module sends the input voice to a keyword matching part in the central processing module;
(4) a matching unit in the keyword matching part sequentially searches input voices closest to the input voices in a keyword library, a local dialect library, a fuzzy voice library and a personal voice library according to preset priority, and takes the preset voices as the standard when the matching degree of the input voices and the preset voices reaches more than 80%; when the matching degree of the input voice and the preset voice is less than 80%, multiplying and adding the key word library, the local dialect library, the fuzzy voice library and the personal voice library with respective correlation degrees to form synthesized voice, and verifying the synthesized voice by the matching unit again; recording the correlation degree of the input voice, the keyword library, the local dialect library, the fuzzy voice library and the personal voice library once voice matching is completed, and adjusting the priority before the next voice matching according to the average value of the correlation degree recorded for many times;
(5) the physical sign type selection part selects a corresponding physical sign type in the physical sign type configuration library according to the determined matching voice;
(6) the numerical integrity judgment part judges the numerical integrity of the input physical sign parameters, and sends out voice example prompts when the numerical integrity judgment part does not meet a numerical integrity threshold; for example, when the input physical sign parameter is blood pressure, when the user inputs only one of high pressure and low pressure through voice, and the numerical integrity judgment part judges that the numerical integrity judgment part does not meet the numerical integrity threshold, a voice example prompt is sent to remind the user to continue inputting voice;
(7) a numerical value abnormity determining unit in the numerical value abnormity degree determining part performs numerical value abnormity determination on the input physical sign parameters, and after the numerical value abnormity determining unit determines that the input voice is abnormal, information is sent to a retest reminding part to remind a user to perform voice input again; the case of determining the input voice abnormality includes: letters, Chinese or English are mixed in the process of inputting the voice, and the numerical value of the input voice is obviously abnormal (for example, the height/weight inputs a numerical value of four digits), and the like.
(8) The threshold value judging part is used for carrying out threshold value judgment on the input physical sign parameters according to a preset threshold value; when the input physical sign parameters are within a normal qualified range, simultaneously displaying the threshold value and the input physical sign parameters to a user; when the input physical sign parameter exceeds a first threshold but does not exceed a second threshold, simultaneously displaying the threshold and the input physical sign parameter to a user, and sending low early warning level information to an early warning level selection part; when the input physical sign parameters exceed the second threshold, simultaneously displaying the threshold and the input physical sign parameters to the user, and sending high early warning level information to an early warning level selection part;
(9) the early warning level selection part determines an early warning level according to the part of the input voice exceeding the threshold value, and when the input physical sign parameter exceeds the first threshold value but does not exceed the second threshold value, the early warning level selection part selects and determines a low early warning level and sends out warning information to a user in time through the voice output module; when the input physical sign parameters exceed a second threshold value, selecting and determining a high early warning level, and sending warning information to a user in time through a voice output module;
(10) the health care knowledge judging part selects corresponding health care knowledge from the health care knowledge base according to the type of the input voice exceeding the threshold value;
(11) and playing the corresponding health care knowledge to the user through the voice output module.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.
Claims (6)
1. A sign monitoring system through voice input comprises a voice input module, a central processing module and a voice output module, wherein the voice input module is connected with the central processing module, and the central processing module is connected with the voice output module; the method is characterized in that: the central processing module is integrated with a voice recognition module, a judgment module and an early warning module;
the voice recognition module comprises a keyword matching part, a physical sign type selection part and a physical sign type configuration library;
the judgment module comprises a numerical value integrity judgment part, a numerical value abnormality judgment part and a retest reminding part;
the early warning module comprises a threshold judging part, an early warning level selecting part, a health care knowledge judging part and a health care knowledge base;
the keyword matching part is provided with a matching unit, a keyword library, a local dialect library, a fuzzy voice library and a personal voice library; the matching unit searches relevant words of input voice from a keyword library, a local dialect library, a fuzzy voice library and a personal voice library and finally forms matched voice;
the keyword matching part is also provided with a priority judging unit and a correlation judging unit, the priority judging unit automatically forms a matching unit according to a preset result and a later-stage adjusting result to search the priority level of the related vocabulary of the input voice from the keyword library, the local dialect library, the fuzzy voice library and the personal voice library, the correlation judging unit forms the correlation with the keyword library, the local dialect library, the fuzzy voice library and the personal voice library according to the result of finally forming the matched voice each time, and the later-stage adjusting is carried out on the priority judging unit according to the correlation;
the threshold determination unit performs threshold determination on the input voice according to a preset threshold; the early warning level selection part determines the early warning level according to the part of the input voice exceeding the threshold value, and the health care knowledge judgment part selects the corresponding health care knowledge according to the type of the input voice exceeding the threshold value.
2. The signs monitoring system through voice entry of claim 1, wherein: the physical sign types in the physical sign type selection part comprise blood pressure, heart rate, body temperature, height/weight and blood sugar information.
3. The signs monitoring system through voice entry of claim 1, wherein: the physical sign type configuration library can be increased or decreased through an external device, and the external device comprises a mobile hard disk, a mobile phone, a watch, an ipad or a computer which can be connected to a network.
4. The signs monitoring system through voice entry of claim 1, wherein: the numerical integrity judgment part is provided with a numerical integrity threshold value and a voice example prompt, and the voice example prompt is sent out when the voice input does not accord with the numerical integrity threshold value.
5. The signs monitoring system through voice entry of claim 1, wherein: and a numerical value abnormity judging unit is arranged in the numerical value abnormity degree judging part, and after the numerical value abnormity judging unit judges that the input voice is abnormal, information is sent to a retesting reminding part to remind a user of carrying out voice input again.
6. The signs monitoring system through voice entry of claim 1, wherein: the health care knowledge base is connected with the hospital background and is synchronous with the hospital background to update the health care knowledge base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810131046.5A CN110136725B (en) | 2018-02-09 | 2018-02-09 | Sign monitoring system through voice input |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810131046.5A CN110136725B (en) | 2018-02-09 | 2018-02-09 | Sign monitoring system through voice input |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110136725A CN110136725A (en) | 2019-08-16 |
| CN110136725B true CN110136725B (en) | 2021-06-18 |
Family
ID=67567852
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810131046.5A Active CN110136725B (en) | 2018-02-09 | 2018-02-09 | Sign monitoring system through voice input |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110136725B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115100817A (en) * | 2022-06-13 | 2022-09-23 | 深圳市牛角尖大数据软件有限公司 | Blood glucose monitoring system based on intelligent watch |
| CN117594178B (en) * | 2024-01-17 | 2024-03-29 | 江苏曼荼罗软件股份有限公司 | Method, system, storage medium and equipment for processing vital sign information |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003263503A (en) * | 2002-03-08 | 2003-09-19 | Matsushita Electric Ind Co Ltd | Health care system |
| CN103646366A (en) * | 2013-11-15 | 2014-03-19 | 北京耀华康业科技发展有限公司 | Interactive type autonomous heath management system and method |
| CN105395168A (en) * | 2015-10-23 | 2016-03-16 | 国家体育总局体育科学研究所 | Determination method for body fat content of Chinese adults |
| CN105411545A (en) * | 2015-12-18 | 2016-03-23 | 安徽寰智信息科技股份有限公司 | Health behavior guiding system |
| CN105872687A (en) * | 2016-03-31 | 2016-08-17 | 乐视控股（北京）有限公司 | Method and device for controlling intelligent equipment through voice |
| CN106529120A (en) * | 2016-09-28 | 2017-03-22 | 云诺维康（深圳）科技有限公司 | Health management method and device |
| CN107591156A (en) * | 2017-10-10 | 2018-01-16 | 杭州嘉楠耘智信息科技股份有限公司 | Audio recognition method and device |
-
2018
- 2018-02-09 CN CN201810131046.5A patent/CN110136725B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003263503A (en) * | 2002-03-08 | 2003-09-19 | Matsushita Electric Ind Co Ltd | Health care system |
| CN103646366A (en) * | 2013-11-15 | 2014-03-19 | 北京耀华康业科技发展有限公司 | Interactive type autonomous heath management system and method |
| CN105395168A (en) * | 2015-10-23 | 2016-03-16 | 国家体育总局体育科学研究所 | Determination method for body fat content of Chinese adults |
| CN105411545A (en) * | 2015-12-18 | 2016-03-23 | 安徽寰智信息科技股份有限公司 | Health behavior guiding system |
| CN105872687A (en) * | 2016-03-31 | 2016-08-17 | 乐视控股（北京）有限公司 | Method and device for controlling intelligent equipment through voice |
| CN106529120A (en) * | 2016-09-28 | 2017-03-22 | 云诺维康（深圳）科技有限公司 | Health management method and device |
| CN107591156A (en) * | 2017-10-10 | 2018-01-16 | 杭州嘉楠耘智信息科技股份有限公司 | Audio recognition method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110136725A (en) | 2019-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9711139B2 (en) | Method for building language model, speech recognition method and electronic apparatus | |
| US7392187B2 (en) | Method and system for the automatic generation of speech features for scoring high entropy speech | |
| CN103578467B (en) | Acoustic model building method, voice recognition method and electronic device | |
| US6839667B2 (en) | Method of speech recognition by presenting N-best word candidates | |
| US7840404B2 (en) | Method and system for using automatic generation of speech features to provide diagnostic feedback | |
| US9613621B2 (en) | Speech recognition method and electronic apparatus | |
| US9483459B1 (en) | Natural language correction for speech input | |
| US7853446B2 (en) | Generation of codified electronic medical records by processing clinician commentary | |
| US9704483B2 (en) | Collaborative language model biasing | |
| EP3504709B1 (en) | Determining phonetic relationships | |
| US20190080687A1 (en) | Learning-type interactive device | |
| CN111833853B (en) | Voice processing method and device, electronic equipment and computer readable storage medium | |
| CN109637537B (en) | Method for automatically acquiring annotated data to optimize user-defined awakening model | |
| EP3736807A1 (en) | Apparatus for media entity pronunciation using deep learning | |
| CN112417128B (en) | Method and device for recommending dialect, computer equipment and storage medium | |
| US7769592B2 (en) | Automatic selection of a disambiguation data field for a speech interface | |
| CN104008752A (en) | Speech recognition device and method, and semiconductor integrated circuit device | |
| KR20190112962A (en) | Cognitive rehabilitation training system | |
| US9805740B2 (en) | Language analysis based on word-selection, and language analysis apparatus | |
| CN109872714A (en) | A kind of method, electronic equipment and storage medium improving accuracy of speech recognition | |
| CN110136725B (en) | Sign monitoring system through voice input | |
| JP2004094257A (en) | Method and apparatus for generating question of decision tree for speech processing | |
| CN105786204A (en) | Information processing method and electronic equipment | |
| CN113935331A (en) | Abnormal semantic truncation detection method, device, equipment and medium | |
| US8401855B2 (en) | System and method for generating data for complex statistical modeling for use in dialog systems |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 212000 room 314, floor 3, building 35, Xuefu Huating, No. 56, Xuefu Road, Zhenjiang City, Jiangsu Province Patentee after: Jiangsu Yitong Intelligent Technology Co.,Ltd. Address before: 212000 room 314, 3rd floor, building 35, Xuefu Huating, No.56 Xuefu Road, Jingkou District, Zhenjiang City, Jiangsu Province Patentee before: JIANGSU ETOWN DIGITAL TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address |