CN104851424A - Voice recognition caring and nursing method and system - Google Patents
Voice recognition caring and nursing method and system Download PDFInfo
- Publication number
- CN104851424A CN104851424A CN201510158160.3A CN201510158160A CN104851424A CN 104851424 A CN104851424 A CN 104851424A CN 201510158160 A CN201510158160 A CN 201510158160A CN 104851424 A CN104851424 A CN 104851424A
- Authority
- CN
- China
- Prior art keywords
- voice signal
- care
- nursing
- speech recognition
- judge
- 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
Landscapes
- Accommodation For Nursing Or Treatment Tables (AREA)
Abstract
The invention provides a voice recognition caring and nursing method and system. The voice recognition caring and nursing system includes a host, at least one submachine and a caring and nursing unit, the submachines are installed in a room and detect plural acoustical signals existing in an environment at the same time, and whether an acoustical signal has a fundamental frequency-overtone structure is utilized to judge whether the acoustical signal is a voice signal or a non-voice signal; if the acoustical signal is a voice signal, content of words and expressions of the voice signal is recognized, whether one of preset plural help-seeking messages is contained is judged, and if the acoustical signal is a non-voice signal, whether the acoustical signal is one of preset plural abnormal sounds is judged; and if the content of words and expressions of the voice signal is a help-seeking message, or the non-voice signal is an abnormal sound, the host sends out warning information to the caring and nursing unit, thereby notifying a family member or a nurse to come to check.
Description
Technical field
The present invention is the technology about a kind of house care and nursing, refers to a kind of method and system thereof of speech recognition care and nursing especially.
Technical background
Along with China is about to march toward aging society, medical care nursing quality is directly social concern with nurse shortage of manpower, People or handicapped person, to suffer from the be in security of daily life of heavy patient be the key factor affecting quality of life, therefore in the hypodynamic situation of present stage ward, how to maintain the home safety needing care and nursing object, making them can obtain succouring when there is emergency is the important problem of item.
Current part home for destitute or medical institutes be between every room mounting monitor to look after the situation of the old men, image-type and sound type two kinds can be divided into for the monitor of long distance nurse, wherein image-type uses the monitor of closed-circuit television or network communicating system to gather image, make supvr and care and nursing person can see the movement of all personnel in monitoring range, advantage is that all situations in monitoring range is all very clear, but shortcoming is that monitor has dead angle, and need people and look at supervision image always, in addition image-type monitor is comparatively useful in infant's care and nursing, for other age levels by care and nursing person shortage privacy.And the current rarely seen vagitus that is used on the market of sound type monitoring system is detected, People or care and nursing that is handicapped, that suffer from heavy patient cannot be applied in.
The present invention is directed to the disappearance of above-mentioned technology, propose a kind of method and system thereof of speech recognition care and nursing, effectively to overcome above-mentioned problem.
Summary of the invention
Fundamental purpose of the present invention is providing a kind of method and system thereof of speech recognition care and nursing, its be room installing main frame or handset to monitor sound, judge that sound is voice or non-voice, distress signals whether are comprised or whether non-voice is urgent abnormal noise in further analyzing speech, to determine to start rescue mechanism the need of sending information warning, monitoring sound is without dead angle, and comparatively monitoring image is more accurate.
Another object of the present invention is providing a kind of method of speech recognition care and nursing, and it utilizes in the sound spectrum of voice signal whether to have fundamental frequency-general frequency structure to judge that whether voice signal is for voice.
The object of the present invention is to provide a kind of method of speech recognition care and nursing, it is stored in audio database by the feature of representative distress signals or abnormal noise, the distress signals or abnormal noise that analyze store in advance if having in voice signal, representative occurs outside may having a mind to, then send information warning.
In order to achieve the above object, the invention provides a kind of method of speech recognition care and nursing system, judge that whether environmental aspect is urgent by speech recognition care and nursing system, comprise the following steps: the plural voice signal in speech recognition care and nursing systems scan environment; Fundamental frequency-general frequency structure decision voice signal is utilized to be voice signal or non-speech audio; If voice signal is voice signal, then the words and expressions content of identification voice signal, judges whether the wherein a kind of of the plural distress signals comprising acquiescence, and if voice signal is non-speech audio, then judge that whether it is the wherein a kind of of default plural number kind abnormal noise; And if the words and expressions content of voice signal is distress signals, or non-speech audio is abnormal noise, then speech recognition care and nursing system external sends information warning.
The present invention separately provides a kind of speech recognition care and nursing system, comprise main frame, at least one handset and care and nursing unit, plural voice signal in handset detecting environment, main frame is connected in a wired or wireless manner with handset, after receiving voice signal, whether there is fundamental frequency-general frequency structure to determine whether voice signal by inspection voice signal, judging that the content of voice signal is the need of sending information warning further; Care and nursing unit is connected in a wired or wireless manner with main frame, if information warning sent by main frame, then care and nursing unit starts rescue mechanism, such as, set out nurse and go scene is actual visits whether someone needs relief.
The method of the speech recognition care and nursing that advantage of the present invention is to provide and system thereof install main frame or handset in a room to monitor sound, between every, estrade machine detecting sound is installed in room, be voice or non-voice by host analysis voice signal again, distress signals or abnormal noise whether is comprised in further analysis voice signal, to determine to start rescue mechanism the need of sending information warning, by judging that voice signal is voice signal and analyzes content wherein, speech recognition care and nursing system of the present invention more can be made to be used in People, the care and nursing of handicapped person or severe case, make sound type monitoring system no longer only for detecting vagitus, and can to being taked response by the emergency of care and nursing person, and by monitoring sound without dead angle, comparatively monitor image more accurate.
Accompanying drawing explanation
Fig. 1 is the calcspar of speech recognition care and nursing system of the present invention.
Fig. 2 is the calcspar of main frame internal mode blocking in speech recognition care and nursing system of the present invention.
Fig. 3 is the configuration diagram of handset in speech recognition care and nursing system of the present invention.
Fig. 4 is the process flow diagram of the care and nursing method of speech recognition care and nursing of the present invention.
Embodiment
The present invention is a kind of method and system thereof of speech recognition care and nursing, it is the voice signal utilizing microphone array to detect indoor, room, and voice signal is analyzed, determine whether voice, if voice then identification speak content whether comprise distress signals then start rescue mechanism; If not voice then determine whether to represent the unexpected abnormal noise occurred, start rescue mechanism, by the characteristic of sound without dead angle, make monitoring function more powerful than video recording monitoring.
Fig. 1 is the calcspar of speech recognition care and nursing system provided by the present invention, comprise main frame 10, at least one handset 12 and care and nursing unit 14, handset 12 is installed in the interior space in room and so on, in order to detect the plural voice signal in environment, main frame 10 and handset 12 are connected in a wired or wireless manner, as grid line or telephone wire, after handset 12 detects voice signal and sends main frame 10 reception to, main frame 10 first can determine whether the voice signal of voice, judge that the content of voice signal is the need of sending information warning more further, such as voice why speak content, if not whether voice are then dangerous abnormal noise, care and nursing unit 14 is connected in a wired or wireless manner with main frame 10, as grid line, telephone wire or wireless network transmissions, this care and nursing unit 14 can be mansion and saves personnel, neighbours, older family, neighbouring medical institutes or emergency unit etc. from damage, if information warning sent by main frame 10, then care and nursing unit starts rescue mechanism, and such as neighbours or nurse can go scene is actual visits whether someone needs relief.
Fig. 2 is main frame internal mode blocking calcspar, main frame 10 comprises sound collection unit 102, audio database 104, acoustic contrast's analytic unit 106 and acoustic recognition unit 108, wherein sound collection unit 102 stores the voice signal that sends of handset and the collection of care and nursing person voice data in order to record, plural distress signals and the abnormal noise of acquiescence is stored in audio database 104, such as save somebody's life, article drop clash etc. on the ground, acoustic contrast's analytic unit 106 will in order to judge whether voice signal is voice signal, it is in the sound spectrum of voice signal, check whether fundamental frequency-general frequency structure has the fundamental frequency of 1001200 hertz, if have, reexamine in voice signal and whether there is high frequency multiplication, have, this voice signal is sorted out the voice signal for people sends, the method can be applicable to the environment of more than noise ratio 5dB, be enough to be useful in general indoor environment, acoustic recognition unit 108 is after when acoustic contrast's analytic unit 106, to analyze voice signal be voice signal, a kind of speech recognition method is utilized to go to judge the words and expressions content of voice signal, this speech recognition method is current existing speech recognition software or phonetic recognition algorithm, identification method as hybrid in SiriOus, K-means, vectorial support machine, hiding markov model, or the speech recognition software etc. that Window7 is built-in.
Fig. 3 is the configuration diagram of handset, microphone array 122 is provided with outside handset 12, in order to detect the sound in environment, and the source position of distinguishable sound, monitoring means 124 is provided with to receive this voice signal in handset 12, and send main frame to by handset 12 and the online of main frame, in addition handset 12 is more provided with loudspeaker 126, microphone array 122 and loudspeaker 126 are used as intercom, when crying for help or uncommon sound is sent to care and nursing unit by main frame, the personnel of care and nursing unit can inquire whether what situation occurs, and the inquiry of care and nursing personnel can be sent by the space of loudspeaker 126 at handset 12 place.
Fig. 4 is the process flow diagram of the care and nursing method of speech recognition care and nursing of the present invention, and the handset first in step S10 speech recognition care and nursing system utilizes microphone array to detect simultaneous plural voice signal in environment; Step S12 voice signal is sent to main frame, and whether the sound spectrum of host computer voice signal comprises fundamental frequency-general frequency structure to judge that whether voice signal is for voice signal, and fundamental frequency-general frequency structure meets the fundamental frequency of sentence 1001200 hertz and has the characteristic of high frequency multiplication; The content of voice signal is differentiated in step S14, if voice signal is voice signal, then utilize the words and expressions content of existing speech recognition method identification voice signal, judge whether to comprise the wherein a kind of of the plural distress signals of acquiescence in data bank, if and voice signal is non-speech audio, then judge whether it is the wherein a kind of of the plural number kind abnormal noise of acquiescence in audio database; Last as described in step S16, if the words and expressions content of voice signal comprises distress signals, " help " two word is comprised as having in voice signal, or non-speech audio is abnormal noise, boiled sound, birdie, the glass breaking sound of such as boiling water, cough ouch sound etc., then the speech recognition care and nursing system external of main frame sends information warning to care and nursing unit, to notify that care and nursing personnel (as neighbours, nurse) inquire or go to scene to watch.
As described above, be only preferred embodiment of the present invention, is not used for limiting scope of the invention process.All equalizations of doing according to the feature described in the present patent application scope and spirit change or modify, and all should be included in claim of the present invention.
Claims (6)
1. a method for speech recognition care and nursing, judge that whether environmental aspect is urgent by speech recognition care and nursing system, the method comprises the following steps:
Simultaneous plural voice signal in this speech recognition care and nursing systems scan environment;
Check whether this voice signal has fundamental frequency-general frequency structure, judge that this voice signal is voice signal or non-speech audio;
If this voice signal is this voice signal, the then words and expressions content of this voice signal of identification, judge whether the wherein a kind of of the plural distress signals comprising acquiescence, and if this voice signal be this non-speech audio, then judge that whether it is the wherein one of default plural number kind abnormal noise;
If the words and expressions content of this voice signal be these distress signals wherein or this non-speech audio be this abnormal noise wherein, then this speech recognition care and nursing system external sends information warning.
2. the method for the speech recognition care and nursing according to claims 1, wherein this voice signal judges words and expressions content by speech recognition method.
3. the method for the speech recognition care and nursing according to claims 1, wherein this judges that the step of this voice signal is in the sound spectrum of this voice signal, check whether this fundamental frequency-general frequency structure has the fundamental frequency of 100 1200 hertz, if have, then reexamine in this voice signal and whether have high frequency multiplication, if having, this voice signal is classified as this voice signal.
4. the method for the speech recognition care and nursing according to claims 1, wherein these distress signals and this abnormal noise are stored in audio database; Wherein this voice signal is told as after this non-speech audio, compare with this abnormal noise in this audio database, if this non-speech audio has with this abnormal noise the person of conforming to, then judges this non-speech audio as this abnormal noise wherein; After wherein the words and expressions content of this voice signal is recognized, compare with these distress signals in this audio database, have with these distress signals the person of conforming to if having in this voice signal, then judge that the words and expressions content of this voice signal comprises these distress signals wherein; Wherein this environment is the interior space; Wherein this speech recognition care and nursing system is the source position utilizing microphone array to judge this voice signal; Wherein this speech recognition care and nursing system sends this information warning with grid line, telephone wire or wireless transmission method.
5.-kind of speech recognition care and nursing system, comprising:
At least one handset, the plural voice signal in detecting environment;
Main frame, be connected in a wired or wireless manner with this at least one handset, receive this voice signal, whether there is fundamental frequency-general frequency structure to judge that whether this voice signal is for voice signal by this voice signal of inspection, the content of further this voice signal of judgement is the need of sending alert not information, and care and nursing unit, is connected in a wired or wireless manner with this main frame, receive this information warning, start rescue mechanism.
6. the speech recognition care and nursing system according to claims 5, wherein this at least one handset is, with network circuit, telephone wire or wireless transmission method, this voice signal is sent to this main frame; Wherein this main frame sends this information warning with grid line, telephone wire or wireless transmission method; Wherein more comprise sound collection unit in this main frame or this at least one handset, store this voice signal to record; Wherein comprise acoustic contrast's analytic unit in this main frame, in order to judge whether this voice signal is this voice signal; Wherein comprise acoustic recognition unit in this main frame, utilize speech recognition method to judge the words and expressions content of this voice signal; Wherein comprise audio database in this main frame, in order to store plural distress signals and plural abnormal noise; Wherein this at least one handset more comprises microphone array to receive this voice signal and to judge the source position of this voice signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510158160.3A CN104851424A (en) | 2015-04-06 | 2015-04-06 | Voice recognition caring and nursing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510158160.3A CN104851424A (en) | 2015-04-06 | 2015-04-06 | Voice recognition caring and nursing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104851424A true CN104851424A (en) | 2015-08-19 |
Family
ID=53851032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510158160.3A Pending CN104851424A (en) | 2015-04-06 | 2015-04-06 | Voice recognition caring and nursing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104851424A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105336325A (en) * | 2015-09-25 | 2016-02-17 | 百度在线网络技术（北京）有限公司 | Speech signal recognition and processing method and device |
| CN108053824A (en) * | 2017-12-11 | 2018-05-18 | 河海大学文天学院 | A kind of patrol rescue unmanned plane and its method of work based on voice signal |
| CN109785581A (en) * | 2019-02-26 | 2019-05-21 | 珠海格力电器股份有限公司 | A kind of monitoring method, device, storage medium and equipment |
| CN110300997A (en) * | 2017-02-16 | 2019-10-01 | 松下知识产权经营株式会社 | Abnormal conditions notify system and abnormal conditions notification method |
| CN110969805A (en) * | 2018-09-30 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | Safety detection method, device and system |
| CN112331208A (en) * | 2020-09-30 | 2021-02-05 | 音数汇元(上海)智能科技有限公司 | Personal safety monitoring method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0033412A2 (en) * | 1980-01-23 | 1981-08-12 | Scott Instruments Company | Method and apparatus for speech recognition |
| CN102521945A (en) * | 2011-12-02 | 2012-06-27 | 无锡奥盛信息科技有限公司 | Calling detection alarming method and device |
| CN103543814A (en) * | 2012-07-16 | 2014-01-29 | 瑞昱半导体股份有限公司 | Signal processing device and signal processing method |
| CN104269016A (en) * | 2014-09-22 | 2015-01-07 | 北京奇艺世纪科技有限公司 | Alarm method and device |
-
2015
- 2015-04-06 CN CN201510158160.3A patent/CN104851424A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0033412A2 (en) * | 1980-01-23 | 1981-08-12 | Scott Instruments Company | Method and apparatus for speech recognition |
| CN102521945A (en) * | 2011-12-02 | 2012-06-27 | 无锡奥盛信息科技有限公司 | Calling detection alarming method and device |
| CN103543814A (en) * | 2012-07-16 | 2014-01-29 | 瑞昱半导体股份有限公司 | Signal processing device and signal processing method |
| CN104269016A (en) * | 2014-09-22 | 2015-01-07 | 北京奇艺世纪科技有限公司 | Alarm method and device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105336325A (en) * | 2015-09-25 | 2016-02-17 | 百度在线网络技术（北京）有限公司 | Speech signal recognition and processing method and device |
| CN110300997A (en) * | 2017-02-16 | 2019-10-01 | 松下知识产权经营株式会社 | Abnormal conditions notify system and abnormal conditions notification method |
| CN108053824A (en) * | 2017-12-11 | 2018-05-18 | 河海大学文天学院 | A kind of patrol rescue unmanned plane and its method of work based on voice signal |
| CN110969805A (en) * | 2018-09-30 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | Safety detection method, device and system |
| CN109785581A (en) * | 2019-02-26 | 2019-05-21 | 珠海格力电器股份有限公司 | A kind of monitoring method, device, storage medium and equipment |
| CN112331208A (en) * | 2020-09-30 | 2021-02-05 | 音数汇元(上海)智能科技有限公司 | Personal safety monitoring method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105489217A (en) | Method and system of monitoring the aged through sound by means of internet of things | |
| US10506411B1 (en) | Portable home and hotel security system | |
| CN104851424A (en) | Voice recognition caring and nursing method and system | |
| US10176705B1 (en) | Audio monitoring and sound identification process for remote alarms | |
| US20230334973A1 (en) | System and method for generating an alert based on noise | |
| TW201342313A (en) | Voice recognition nursing method and system thereof | |
| MX2007008766A (en) | Management and assistance system for the deaf. | |
| CN102881129A (en) | Alarm receiving host, system and method | |
| US20190180735A1 (en) | Ambient sound classification based message routing for local security and remote internet query systems | |
| KR102066225B1 (en) | Smart health care apparatus, system and method using artificial intelligence | |
| US20100052903A1 (en) | Voice recorder based position registration | |
| KR20180093596A (en) | Safety Sensor Based on Scream Detection | |
| KR101342448B1 (en) | Remote monitoring apparatus and the method thereof | |
| US20190228628A1 (en) | Audio monitoring system | |
| JP2006235996A (en) | Warning system and warning device | |
| KR102488741B1 (en) | Emergency bell system with improved on-site situation identification | |
| JP5627962B2 (en) | Anomaly detection device | |
| JP2005309965A (en) | Home security device | |
| CN218273143U (en) | Digital twin monitoring system | |
| KR20200047083A (en) | Emergency monitoring system for people with disabilities and the elderly according to voice recognition | |
| JP2007228287A (en) | Intercom device | |
| KR20030027925A (en) | Rescue request apparatus using voice recognition and method thereof | |
| US20230370302A1 (en) | System, method and apparatus for propagating a primary alert of a monitoring system or device | |
| JP2001307238A (en) | System for preventing crime and method for detecting cutting sound | |
| JP2012083956A (en) | Security device, security system and security method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| CB03 | Change of inventor or designer information |
Inventor after: He Tao Inventor after: Liang Zhidong Inventor after: Ye Xinquan Inventor after: Lin Wuwei Inventor after: Pang Jihong Inventor before: He Tao Inventor before: Wang Xiaoyang |
|
| COR | Change of bibliographic data | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160426 Address after: 325000 Zhejiang, Ouhai, South East Road, No. 38, Wenzhou National University Science Park Incubator Applicant after: WENZHOU UNIVERSITY OUJIANG COLLEGE Address before: 325005, No. 128-66, Civil Aviation Road, PU shoe city, Lucheng District, Zhejiang, Wenzhou Applicant before: He Tao |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150819 |
|
| WD01 | Invention patent application deemed withdrawn after publication |