CN103198838A - Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system - Google Patents
Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system Download PDFInfo
- Publication number
- CN103198838A CN103198838A CN2013101063060A CN201310106306A CN103198838A CN 103198838 A CN103198838 A CN 103198838A CN 2013101063060 A CN2013101063060 A CN 2013101063060A CN 201310106306 A CN201310106306 A CN 201310106306A CN 103198838 A CN103198838 A CN 103198838A
- Authority
- CN
- China
- Prior art keywords
- sound
- audio frequency
- frequency characteristics
- short
- term
- 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
Images
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The invention discloses an abnormal sound monitoring method and an abnormal sound monitoring device used for an embedded system. The monitoring method includes the following steps: (1) obtaining audio data, framing the audio data with a Hamming window, calculating the short-time magnitude and the short-time super-dynamic threshold rate of each frame of sound, and obtaining effective sound segments through judgment according to the short-time magnitude and the short-time super-dynamic threshold rate of each frame of sound; (2) building an audio characteristic template base by utilization of a plurality of sound samples of different kinds, wherein the sound samples of the same kind correspond to a template in the audio characteristic template base; and (3) extracting audio characteristic parameters of the effective sound segments obtained in the step (1), matching the audio characteristic parameters with templates in the audio characteristic template base and determining categories of the effective sound segments. The monitoring method can achieve real-time sampling analysis of audio signals and can effectively identify abnormal sound in a monitored place, the identification speed is high, and moreover the monitoring device is simple in structure and strong in reliability.
Description
Technical field
The present invention relates to electronic device field, be specifically related to a kind of abnormal sound method for supervising and supervising device for embedded system.
Background technology
Along with the expansion day by day that increases, monitors scale rapidly of video monitoring demand, the artificial supervision can not be satisfied monitoring requirement far away, and " intellectuality " of video monitoring system becomes more and more urgent.
At present, intelligent monitoring is mainly based on vision signal, by the image sequence of video camera recording is analyzed automatically, realization is to the real-time monitoring of unusual condition in the dynamic scene, but video analysis exists calculated amount big, realize complicated, the characteristics that system performance is had relatively high expectations.
In some monitoring place, sound signal has often comprised than the more direct clue of vision signal, except normal ambient sound, also have a class sound can effectively disclose abnormality and burst accident, be referred to as abnormal sound, typical abnormal sound comprises the broken sound of shot, explosive sound and glass etc. in the monitoring place.By can finish the real-time monitoring to unusual condition equally to the automatic analysis of sound signal, and than video monitoring system, Audio Monitoring System based on voice recognition possesses lower complexity and the counting yield of Geng Gao, realizes in embedded system easilier.
Granted publication number be CN 101494049 B disclosure of the Invention a kind of audio frequency characteristics Parameter Extraction method for Audio Monitoring System, operation steps is as follows: (1) is carried out the branch frame and it is divided into the voice data frame sequence the audio sample burst according to the duration of the short time of setting; (2) respectively to short-time energy, short-time zero-crossing rate and the short-term information entropy of each this frame of audio data frame extract real-time in this voice data frame sequence; (3) the voice data frame sequence is carried out segmentation and it is divided into the short section of audio frequency sequence; And on the above-mentioned parameter basis, extract time domain and the frequency domain character of all audio data frames in the short section of comprehensive each audio frequency and take into full account its former and later two constantly between the characteristic parameter of audio frequency difference: frequency difference during the short section of audio frequency is used for Audio Monitoring System.
Based on the embedded system self characteristics, the algorithm of voice recognition can not be too complicated, require the voice recognition accuracy rate to want high simultaneously, in a certain specific monitoring place, the kind of abnormal sound can be predicted in advance, it is relatively low to set computational complexity accordingly, and can reach the purpose of quick and precisely carrying out voice recognition.
Summary of the invention
The invention provides a kind of abnormal sound method for supervising and supervising device for embedded system, can realize the real-time sampling analysis of sound signal, can identify the abnormal sound in the monitoring place effectively, recognition speed is fast, and supervising device is simple in structure, good reliability.
A kind of abnormal sound method for supervising for embedded system may further comprise the steps:
(1) obtain voice data, adopt Hamming window that this voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment.
Dynamic threshold has reacted the average amplitude level of ground unrest, crosses the dynamic threshold rate in short-term compared to short-time zero-crossing rate, can effectively reduce the influence of ground unrest.
Effectively the length of acoustic segment is long or too short, can abandon, and the judgement of length can be set as required.
The operation that obtains effective acoustic segment according to the amplitude in short-term of every frame sound and mistake dynamic threshold rate judgement in short-term is as follows:
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all greater than respective threshold, then with the starting point that is designated as sound the finish time of two continuous frames;
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all be not more than respective threshold, then with the cut off that is designated as sound the finish time of two continuous frames.
(2) utilize different types of a plurality of sample sound, set up the audio frequency characteristics template base, wherein, a template in the corresponding audio frequency feature templates of the sample sound of the same kind storehouse.
Utilize different types of a plurality of sample sound, when setting up the audio frequency characteristics template base, the concrete operations of template of calculating a certain kind acoustic phase correspondence are as follows:
2-1, extract the audio frequency characteristics parameter of each sample sound in this kind sound, obtain the audio frequency characteristics argument sequence;
2-2, calculate the average length of the audio frequency characteristics argument sequence of all sample sounds, select to differ minimum sample sound as false form with average length;
2-3, that all the other sample sounds outside the false form are carried out dynamic time bending is regular, obtains the sample sound that audio frequency characteristics argument sequence length all equates;
2-4, calculate the mean value of the audio frequency characteristics parameter of the sample sound after regular, this mean value is the template in the audio frequency characteristics template base.
Can use the sound in the sample sound storehouse when setting up the audio frequency characteristics template base, also can simulate voluntarily, the some kinds of abnormal sounds of selecting to occur at the characteristics of monitoring space carry out the foundation of corresponding template.
Described audio frequency characteristics parameter comprises the first order difference coefficient of amplitude, 12 rank MFCC and 12 rank MFCC in short-term.
Amplitude characterizes the size of a frame wave audio energy in short-term, is important temporal signatures of sound signal, and does not have the demand square operation when calculating, and is conducive to alleviate calculated amount.
MFCC is a kind of characteristic parameter that the generation with human auditory system apperceive characteristic and sound combines, 12 rank MFCC(Mel frequency cepstral coefficients) considered the nonlinear characteristic of human auditory system to have good recognition performance and anti-noise ability.
The first order difference coefficient of 12 rank MFCC has reflected exist between the frame and frame in the audio frequency related, is the time dependent rule of expression audio frequency characteristics parameter, helps to distinguish different voice signals.
The kind of the audio frequency characteristics parameter of extracting in the audio frequency characteristics parameter of extracting when setting up the audio frequency characteristics template base and the sound of the real-time monitoring of collection is the same.
(3) the audio frequency characteristics parameter of the effective acoustic segment that obtains in the extraction step (1), and the template in this audio frequency characteristics parameter and the audio frequency characteristics template base mated, determine the classification of effective acoustic segment.
Concrete operations when the template in this audio frequency characteristics parameter and the audio frequency characteristics template base is mated are as follows:
3-1, calculate the matching distance of each template in each frame sound and audio frequency characteristics template base in effective acoustic segment, obtain the minimum value D of matching distance
Min
3-2, definition distance threshold TH work as D
MinDuring/N≤TH, effectively acoustic segment and D
MinCorresponding template matches is worked as D
Min/ N〉during TH, effectively the arbitrary template in acoustic segment and the audio frequency characteristics template base does not all match, and N is the sound frame number in effective acoustic segment.
As preferably, also comprise step (4), report to the police according to the classification of effective acoustic segment.
The corresponding a kind of abnormal sound of a template in the audio frequency characteristics template base, if effectively the audio frequency characteristics parameter of acoustic segment and the template in the audio frequency characteristics template base are complementary, judge that then this effective acoustic segment is abnormal sound, if abnormal sound, then report to the police, notify the relevant personnel in time to handle, also can report warning message to central server by gigabit ethernet interface.
The present invention pays close attention to the abnormal sound that can disclose unusual condition and burst accident in a certain monitoring place emphatically, limited kind of abnormal sound in this monitoring place can be judged in advance, in a room sudden and violent fried province, the broken sound of glass or shot etc. appear for example, set up corresponding template for each abnormal sound in advance, be stored in the audio frequency characteristics template base, when abnormal sound occurring, only need with the audio frequency characteristics template base in template mate and can identify.
The present invention also provides a kind of abnormal sound supervising device for embedded system, comprises with lower unit:
The audio frequency receiving element is used for obtaining voice data;
Effectively acoustic segment end-point detection unit is used for adopting Hamming window that voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment;
The audio frequency characteristics template base is set up the unit, is used for utilizing different types of a plurality of sample sound, sets up the audio frequency characteristics template base;
The audio feature extraction unit is for the audio frequency characteristics parameter of extracting effective acoustic segment;
Effectively the acoustic segment recognition unit is used for the template of audio frequency characteristics parameter and audio frequency characteristics template base is mated, and determines the classification of effective acoustic segment;
The abnormal alarm unit is used for reporting to the police according to the classification of effective acoustic segment.
The invention provides a kind of abnormal sound method for supervising and supervising device for embedded system, method is simple, can adapt to the real-time sampling analysis of sound signal in a certain specifically monitored space, rapidly the abnormal sound in the identification monitoring place, supervising device is simple in structure, good reliability.
Description of drawings
Fig. 1 is used for the synoptic diagram of the abnormal sound supervising device of embedded system for the present invention;
Fig. 2 is used for the structural representation of the abnormal sound supervising device of embedded system for the present invention;
Fig. 3 is used for the process flow diagram of the abnormal sound method for supervising of embedded system for the present invention;
Fig. 4 is used for the process flow diagram of the abnormal sound method for supervising host process of embedded system for the present invention.
Fig. 5 is used for the process flow diagram of the abnormal sound method for supervising thread of embedded system for the present invention.
Embodiment
Below in conjunction with accompanying drawing, a kind of abnormal sound method for supervising and supervising device for embedded system of the present invention is described in detail.
As shown in Figure 3, a kind of abnormal sound method for supervising for embedded system may further comprise the steps:
(1) obtain voice data, adopt Hamming window that this voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment.
When using Hamming window to carry out the branch frame, window length gets 256, with 16ms(256 sampled point) be a frame, frame displacement 12ms.
The operation that obtains effective acoustic segment according to the amplitude in short-term of every frame sound and mistake dynamic threshold rate judgement in short-term is as follows:
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all greater than respective threshold, then with the starting point that is designated as sound the finish time of two continuous frames;
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all be not more than respective threshold, then with the cut off that is designated as sound the finish time of two continuous frames.
(2) utilize different types of a plurality of sample sound, set up the audio frequency characteristics template base.
The concrete operations of template of calculating a certain kind acoustic phase correspondence are as follows:
2-1, extract the audio frequency characteristics parameter of each sample sound in this kind sound, obtain the audio frequency characteristics argument sequence; The audio frequency characteristics parameter comprises the first order difference coefficient of amplitude, 12 rank MFCC and 12 rank MFCC in short-term;
2-2, calculate the average length of the audio frequency characteristics argument sequence of all sample sounds, select to differ minimum sample sound as false form with average length;
2-3, all the other sample sounds outside the false form are carried out dynamic time bending regular (DTW), obtain the sample sound that audio frequency characteristics argument sequence length all equates;
2-4, calculate the mean value of the audio frequency characteristics parameter of the sample sound after regular, this mean value is the template in the audio frequency characteristics template base.
(3) the audio frequency characteristics parameter of the effective acoustic segment that obtains in the extraction step (1) (comprising the first order difference coefficient of amplitude, 12 rank MFCC and 12 rank MFCC in short-term), and the template in this audio frequency characteristics parameter and the audio frequency characteristics template base mated (adopting and limiting the lattice point matching range is the quick DTW algorithm of diamond-shaped area, referring to: what is strong, what English .Matlab expansion programming [M]. Beijing: publishing house of Tsing-Hua University, 2002-06.), determine the classification of effective acoustic segment.
At first, suppose that effective acoustic segment of a N frame and the matching distance of K template in the audio frequency characteristics template base are respectively D
1, D
2..., D
K, D
MinBe wherein minimum matching distance; (the matching distance algorithm referring to: what is strong, He Ying .Matlab expansion programming [M]. Beijing: publishing house of Tsing-Hua University, 2002-06.)
Secondly, definition distance threshold TH works as D
MinDuring/N≤TH, effectively acoustic segment and D
MinCorresponding template matches is worked as D
Min/ N〉during TH, effectively the arbitrary template in acoustic segment and the audio frequency characteristics template base does not all match.
(4) report to the police according to the classification of effective acoustic segment.
If effectively the masterplate in acoustic segment and the audio frequency characteristics template library is complementary, think that then this effective acoustic segment is abnormal sound, report to the police, otherwise do not report to the police.
Abnormal sound supervising device for embedded system provided by the invention comprises with lower unit:
The audio frequency receiving element is used for obtaining voice data;
Effectively acoustic segment end-point detection unit is used for adopting Hamming window that voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment;
The audio frequency characteristics template base is set up the unit, is used for utilizing different types of a plurality of sample sound, sets up the audio frequency characteristics template base;
The audio feature extraction unit is for the audio frequency characteristics parameter of extracting effective acoustic segment;
Effectively the acoustic segment recognition unit is used for the template of audio frequency characteristics parameter and audio frequency characteristics template base is mated, and determines the classification of effective acoustic segment;
The abnormal alarm unit is used for reporting to the police according to the classification of effective acoustic segment.
The audio frequency receiving element is audio A/D converter; Effective acoustic segment end-point detection unit, audio frequency characteristics template base are set up unit, audio feature extraction unit and effective acoustic segment recognition unit and are integrated in the flush bonding processor, flush bonding processor is done automatic analysis to voice data, flush bonding processor links to each other with audio A/D converter by the I2S bus, flush bonding processor is provided with gigabit ethernet interface, be used for connecting central server, report warning message, structure as shown in Figure 2.
Audio A/D converter adopts 16K sampling rate, 16 sampling precisions, and flush bonding processor adopts built-in Linux operating system.
As shown in Figure 1, abnormal sound supervising device provided by the invention is connected with external microphone by audio input interface, sound signal enters into apparatus of the present invention and carries out real-time voice data analysis, and analysis result is transferred to the central server of far-end by Ethernet, the monitor staff obtains warning message by the distal center server, and carries out corresponding processing scheme.
Sound method for supervising of the present invention adopts multithreading to guarantee real-time, wherein the treatment scheme of host process as shown in Figure 4, specific as follows:
A, audio frequency receiving element read a frame voice data by the I2S bus from audio A/D converter; Judge the current detection state, if quiet section, if execution in step B then is acoustic segment, then execution in step C;
B, currently be in quiet section, according to amplitude in short-term with cross the starting point that the dynamic threshold rate is sought sound in short-term; If detect the starting point of sound, then upgrading detected state is acoustic segment; Come back to steps A;
C, the current acoustic segment that is in are extracted the audio frequency characteristics parameter of this acoustic segment, and according to amplitude in short-term with cross the cut off that the dynamic threshold rate is sought sound in short-term; If detect the cut off of sound, then upgrading detected state is quiet section, judges whether effective acoustic segment length meets the requirements, and then wakes thread up if meet; Come back to steps A;
Wherein the treatment scheme of thread as shown in Figure 5, specific as follows:
1) if waken up execution in step 2 by host process), otherwise continue sleep;
2) template in detected audio frequency characteristics parameter and the audio frequency characteristics template base is compared one by one, if confirm as abnormal sound, then submitted to warning message to give central server by gigabit ethernet interface; Come back to step 1).
Claims (8)
1. an abnormal sound method for supervising that is used for embedded system is characterized in that, may further comprise the steps:
(1) obtain voice data, adopt Hamming window that this voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment;
(2) utilize different types of a plurality of sample sound, set up the audio frequency characteristics template base, wherein, a template in the corresponding audio frequency feature templates of the sample sound of the same kind storehouse;
(3) the audio frequency characteristics parameter of the effective acoustic segment that obtains in the extraction step (1), and the template in this audio frequency characteristics parameter and the audio frequency characteristics template base mated, determine the classification of effective acoustic segment.
2. the abnormal sound method for supervising for embedded system as claimed in claim 1 is characterized in that, also comprises step (4), reports to the police according to the classification of effective acoustic segment.
3. the abnormal sound method for supervising for embedded system as claimed in claim 1 is characterized in that, in the described step (1) according to the amplitude in short-term of every frame sound with cross the judgement of dynamic threshold rate in short-term to obtain the operation of effective acoustic segment as follows:
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all greater than respective threshold, then with the starting point that is designated as sound the finish time of two continuous frames;
If the amplitude in short-term of two continuous frames and in short-term mistake dynamic threshold rate all be not more than respective threshold, then with the cut off that is designated as sound the finish time of two continuous frames.
4. the abnormal sound method for supervising for embedded system as claimed in claim 1, it is characterized in that, utilize different types of a plurality of sample sound in the described step (2), when setting up the audio frequency characteristics template base, the concrete operations of template of calculating a certain kind acoustic phase correspondence are as follows:
2-1, extract the audio frequency characteristics parameter of each sample sound in this kind sound, obtain the audio frequency characteristics argument sequence;
2-2, calculate the average length of the audio frequency characteristics argument sequence of all sample sounds, select to differ minimum sample sound as false form with average length;
2-3, that all the other sample sounds outside the false form are carried out dynamic time bending is regular, obtains the sample sound that audio frequency characteristics argument sequence length all equates;
2-4, calculate the mean value of the audio frequency characteristics parameter of the sample sound after regular, this mean value is the template in the audio frequency characteristics template base.
5. the abnormal sound method for supervising for embedded system as claimed in claim 1 is characterized in that, the concrete operations when in the described step (3) template in this audio frequency characteristics parameter and the audio frequency characteristics template base being mated are as follows:
3-1, calculate the matching distance of each template in each frame sound and audio frequency characteristics template base in effective acoustic segment, obtain the minimum value D of matching distance
Min
3-2, definition distance threshold TH work as D
MinDuring/N≤TH, effectively acoustic segment and D
MinCorresponding template matches is worked as D
Min/ N〉during TH, effectively the arbitrary template in acoustic segment and the audio frequency characteristics template base does not all match, and N is the sound frame number in effective acoustic segment.
6. the abnormal sound method for supervising for embedded system as claimed in claim 1 is characterized in that, described audio frequency characteristics parameter comprises the first order difference coefficient of amplitude, 12 rank MFCC and 12 rank MFCC in short-term.
7. an abnormal sound supervising device that is used for embedded system is characterized in that, comprises with lower unit:
The audio frequency receiving element is used for obtaining voice data;
Effectively acoustic segment end-point detection unit is used for adopting Hamming window that voice data is carried out the branch frame, calculate the amplitude in short-term of every frame sound and cross the dynamic threshold rate in short-term, according to the amplitude in short-term of every frame sound and in short-term the judgement of dynamic threshold rate obtain effective acoustic segment;
The audio frequency characteristics template base is set up the unit, is used for utilizing different types of a plurality of sample sound, sets up the audio frequency characteristics template base;
The audio feature extraction unit is for the audio frequency characteristics parameter of extracting effective acoustic segment;
Effectively the acoustic segment recognition unit is used for the template of audio frequency characteristics parameter and audio frequency characteristics template base is mated, and determines the classification of effective acoustic segment.
8. the abnormal sound supervising device for embedded system as claimed in claim 7 is characterized in that, also comprises: the abnormal alarm unit is used for reporting to the police according to the classification of effective acoustic segment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101063060A CN103198838A (en) | 2013-03-29 | 2013-03-29 | Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101063060A CN103198838A (en) | 2013-03-29 | 2013-03-29 | Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103198838A true CN103198838A (en) | 2013-07-10 |
Family
ID=48721312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101063060A Pending CN103198838A (en) | 2013-03-29 | 2013-03-29 | Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103198838A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104538041A (en) * | 2014-12-11 | 2015-04-22 | 深圳市智美达科技有限公司 | Method and system for detecting abnormal sounds |
| CN104991245A (en) * | 2015-07-29 | 2015-10-21 | 杨珊珊 | Unmanned aerial vehicle early warning apparatus and early warning method thereof |
| CN105203999A (en) * | 2015-10-20 | 2015-12-30 | 陈昊 | Rotorcraft early-warning device and method |
| CN105551491A (en) * | 2016-02-15 | 2016-05-04 | 海信集团有限公司 | Voice recognition method and device |
| CN106328167A (en) * | 2016-08-16 | 2017-01-11 | 成都市和平科技有限责任公司 | Intelligent speech recognition robot and control system |
| CN106448700A (en) * | 2016-10-31 | 2017-02-22 | 济南中维世纪科技有限公司 | High-pitched sound detection method and system |
| EP3147902A1 (en) * | 2015-09-24 | 2017-03-29 | Canon Kabushiki Kaisha | Sound processing apparatus, sound processing method, and computer program |
| CN106887241A (en) * | 2016-10-12 | 2017-06-23 | 阿里巴巴集团控股有限公司 | A kind of voice signal detection method and device |
| CN106971499A (en) * | 2017-04-14 | 2017-07-21 | 北京克路德人工智能科技有限公司 | Intelligent monitor system based on auditory localization |
| CN106971710A (en) * | 2017-03-15 | 2017-07-21 | 国网山东省电力公司威海供电公司 | Electricity business hall personnel voice anomalous event recognition methods and device |
| CN107240405A (en) * | 2017-06-14 | 2017-10-10 | 深圳市冠旭电子股份有限公司 | A kind of audio amplifier and alarm method |
| CN107742517A (en) * | 2017-10-10 | 2018-02-27 | 广东中星电子有限公司 | A kind of detection method and device to abnormal sound |
| CN107786931A (en) * | 2016-08-24 | 2018-03-09 | 中国电信股份有限公司 | Audio-frequency detection and device |
| CN107917342A (en) * | 2017-11-15 | 2018-04-17 | 北京科创三思科技发展有限公司 | The unattended Sound image localization monitoring system and method for natural gas station |
| WO2018068639A1 (en) * | 2016-10-14 | 2018-04-19 | 腾讯科技（深圳）有限公司 | Data recovery method and apparatus, and storage medium |
| CN108447501A (en) * | 2018-03-27 | 2018-08-24 | 中南大学 | Pirate video detection method and system based on audio word under a kind of cloud storage environment |
| CN108694937A (en) * | 2017-04-05 | 2018-10-23 | 陈荣伟 | A kind of method of digital video-audio monitoring system sound intermediate frequency visual control |
| CN108766466A (en) * | 2018-06-19 | 2018-11-06 | 河南孚点电子科技有限公司 | A kind of video control method based on voice signal alarm |
| WO2018228280A1 (en) * | 2017-06-12 | 2018-12-20 | 杭州海康威视数字技术股份有限公司 | Notification information output method, server and monitoring system |
| CN109087655A (en) * | 2018-07-30 | 2018-12-25 | 桂林电子科技大学 | A kind of monitoring of traffic route sound and exceptional sound recognition system |
| CN109357749A (en) * | 2018-09-04 | 2019-02-19 | 南京理工大学 | A kind of power equipment audio signal analysis method based on DNN algorithm |
| CN110148423A (en) * | 2019-04-18 | 2019-08-20 | 广东思派康电子科技有限公司 | A kind of computer readable storage medium and antitheft interactive voice speaker can be used for |
| CN111223486A (en) * | 2019-12-30 | 2020-06-02 | 上海联影医疗科技有限公司 | Alarm device and method |
| CN111354366A (en) * | 2018-12-20 | 2020-06-30 | 沈阳新松机器人自动化股份有限公司 | Abnormal sound detection method and abnormal sound detection device |
| CN111770427A (en) * | 2020-06-24 | 2020-10-13 | 杭州海康威视数字技术股份有限公司 | Microphone array detection method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101067930A (en) * | 2007-06-07 | 2007-11-07 | 深圳先进技术研究院 | Intelligent audio frequency identifying system and identifying method |
| CN101887722A (en) * | 2009-06-18 | 2010-11-17 | 博石金（北京）信息技术有限公司 | Rapid voiceprint authentication method |
| CN102148032A (en) * | 2010-12-03 | 2011-08-10 | 北京声迅电子有限公司 | Abnormal sound detection method and system for ATM (Automatic Teller Machine) |
| CN102163427A (en) * | 2010-12-20 | 2011-08-24 | 北京邮电大学 | Method for detecting audio exceptional event based on environmental model |
| KR20120055090A (en) * | 2010-11-23 | 2012-05-31 | 고려대학교 산학협력단 | Device and method for audio data recognition and system for crime prevention using audio data recognition |
| CN102623009A (en) * | 2012-03-02 | 2012-08-01 | 安徽科大讯飞信息技术股份有限公司 | Abnormal emotion automatic detection and extraction method and system on basis of short-time analysis |
-
2013
- 2013-03-29 CN CN2013101063060A patent/CN103198838A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101067930A (en) * | 2007-06-07 | 2007-11-07 | 深圳先进技术研究院 | Intelligent audio frequency identifying system and identifying method |
| CN101887722A (en) * | 2009-06-18 | 2010-11-17 | 博石金（北京）信息技术有限公司 | Rapid voiceprint authentication method |
| KR20120055090A (en) * | 2010-11-23 | 2012-05-31 | 고려대학교 산학협력단 | Device and method for audio data recognition and system for crime prevention using audio data recognition |
| CN102148032A (en) * | 2010-12-03 | 2011-08-10 | 北京声迅电子有限公司 | Abnormal sound detection method and system for ATM (Automatic Teller Machine) |
| CN102163427A (en) * | 2010-12-20 | 2011-08-24 | 北京邮电大学 | Method for detecting audio exceptional event based on environmental model |
| CN102623009A (en) * | 2012-03-02 | 2012-08-01 | 安徽科大讯飞信息技术股份有限公司 | Abnormal emotion automatic detection and extraction method and system on basis of short-time analysis |
Non-Patent Citations (1)
| Title |
|---|
| 周明忠等: "基于平均幅度和加权过零率的VAD算法及其FPGA实现", 《信息工程大学学报》 * |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104538041A (en) * | 2014-12-11 | 2015-04-22 | 深圳市智美达科技有限公司 | Method and system for detecting abnormal sounds |
| CN104991245A (en) * | 2015-07-29 | 2015-10-21 | 杨珊珊 | Unmanned aerial vehicle early warning apparatus and early warning method thereof |
| US20170092296A1 (en) * | 2015-09-24 | 2017-03-30 | Canon Kabushiki Kaisha | Sound processing apparatus, sound processing method, and storage medium |
| US10109299B2 (en) * | 2015-09-24 | 2018-10-23 | Canon Kabushiki Kaisha | Sound processing apparatus, sound processing method, and storage medium |
| CN106558317A (en) * | 2015-09-24 | 2017-04-05 | 佳能株式会社 | Sound processing apparatus and sound processing method |
| EP3147902A1 (en) * | 2015-09-24 | 2017-03-29 | Canon Kabushiki Kaisha | Sound processing apparatus, sound processing method, and computer program |
| CN105203999A (en) * | 2015-10-20 | 2015-12-30 | 陈昊 | Rotorcraft early-warning device and method |
| CN105551491A (en) * | 2016-02-15 | 2016-05-04 | 海信集团有限公司 | Voice recognition method and device |
| CN106328167A (en) * | 2016-08-16 | 2017-01-11 | 成都市和平科技有限责任公司 | Intelligent speech recognition robot and control system |
| CN107786931A (en) * | 2016-08-24 | 2018-03-09 | 中国电信股份有限公司 | Audio-frequency detection and device |
| US10706874B2 (en) * | 2016-10-12 | 2020-07-07 | Alibaba Group Holding Limited | Voice signal detection method and apparatus |
| CN106887241A (en) * | 2016-10-12 | 2017-06-23 | 阿里巴巴集团控股有限公司 | A kind of voice signal detection method and device |
| WO2018068636A1 (en) * | 2016-10-12 | 2018-04-19 | 阿里巴巴集团控股有限公司 | Method and device for detecting audio signal |
| WO2018068639A1 (en) * | 2016-10-14 | 2018-04-19 | 腾讯科技（深圳）有限公司 | Data recovery method and apparatus, and storage medium |
| CN106448700A (en) * | 2016-10-31 | 2017-02-22 | 济南中维世纪科技有限公司 | High-pitched sound detection method and system |
| CN106448700B (en) * | 2016-10-31 | 2019-09-17 | 济南中维世纪科技有限公司 | A kind of piping detection method and system |
| CN106971710A (en) * | 2017-03-15 | 2017-07-21 | 国网山东省电力公司威海供电公司 | Electricity business hall personnel voice anomalous event recognition methods and device |
| CN108694937A (en) * | 2017-04-05 | 2018-10-23 | 陈荣伟 | A kind of method of digital video-audio monitoring system sound intermediate frequency visual control |
| CN106971499A (en) * | 2017-04-14 | 2017-07-21 | 北京克路德人工智能科技有限公司 | Intelligent monitor system based on auditory localization |
| WO2018228280A1 (en) * | 2017-06-12 | 2018-12-20 | 杭州海康威视数字技术股份有限公司 | Notification information output method, server and monitoring system |
| US11275628B2 (en) | 2017-06-12 | 2022-03-15 | Hangzhou Hikvision Digital Technology Co., Ltd. | Notification information output method, server and monitoring system |
| CN107240405A (en) * | 2017-06-14 | 2017-10-10 | 深圳市冠旭电子股份有限公司 | A kind of audio amplifier and alarm method |
| CN107742517A (en) * | 2017-10-10 | 2018-02-27 | 广东中星电子有限公司 | A kind of detection method and device to abnormal sound |
| CN107917342A (en) * | 2017-11-15 | 2018-04-17 | 北京科创三思科技发展有限公司 | The unattended Sound image localization monitoring system and method for natural gas station |
| CN108447501A (en) * | 2018-03-27 | 2018-08-24 | 中南大学 | Pirate video detection method and system based on audio word under a kind of cloud storage environment |
| CN108447501B (en) * | 2018-03-27 | 2020-08-18 | 中南大学 | Pirated video detection method and system based on audio words in cloud storage environment |
| CN108766466A (en) * | 2018-06-19 | 2018-11-06 | 河南孚点电子科技有限公司 | A kind of video control method based on voice signal alarm |
| CN109087655A (en) * | 2018-07-30 | 2018-12-25 | 桂林电子科技大学 | A kind of monitoring of traffic route sound and exceptional sound recognition system |
| CN109357749A (en) * | 2018-09-04 | 2019-02-19 | 南京理工大学 | A kind of power equipment audio signal analysis method based on DNN algorithm |
| CN111354366A (en) * | 2018-12-20 | 2020-06-30 | 沈阳新松机器人自动化股份有限公司 | Abnormal sound detection method and abnormal sound detection device |
| CN111354366B (en) * | 2018-12-20 | 2023-06-16 | 沈阳新松机器人自动化股份有限公司 | Abnormal sound detection method and abnormal sound detection device |
| CN110148423A (en) * | 2019-04-18 | 2019-08-20 | 广东思派康电子科技有限公司 | A kind of computer readable storage medium and antitheft interactive voice speaker can be used for |
| CN111223486A (en) * | 2019-12-30 | 2020-06-02 | 上海联影医疗科技有限公司 | Alarm device and method |
| CN111223486B (en) * | 2019-12-30 | 2023-02-24 | 上海联影医疗科技股份有限公司 | Alarm device and method |
| CN111770427A (en) * | 2020-06-24 | 2020-10-13 | 杭州海康威视数字技术股份有限公司 | Microphone array detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103198838A (en) | Abnormal sound monitoring method and abnormal sound monitoring device used for embedded system | |
| CN101494049B (en) | Method for extracting audio characteristic parameter of audio monitoring system | |
| EP3522153B1 (en) | Voice control system, wakeup method and wakeup apparatus therefor, electrical appliance and co-processor | |
| CN106714023B (en) | A kind of voice awakening method, system and bone conduction earphone based on bone conduction earphone | |
| CN107928673B (en) | Audio signal processing method, audio signal processing apparatus, storage medium, and computer device | |
| CN102163427B (en) | Method for detecting audio exceptional event based on environmental model | |
| CN108172242B (en) | Improved Bluetooth intelligent cloud sound box voice interaction endpoint detection method | |
| CN103530912A (en) | Attendance recording system having emotion identification function, and method thereof | |
| CN102148032A (en) | Abnormal sound detection method and system for ATM (Automatic Teller Machine) | |
| CN103971681A (en) | Voice recognition method and system | |
| CN108597164B (en) | Anti-theft method, anti-theft device, anti-theft terminal and computer readable medium | |
| WO2020056329A1 (en) | Energy efficient custom deep learning circuits for always-on embedded applications | |
| CN102023604A (en) | Intelligent online monitoring system capable of preventing external damage on transmission line | |
| CN107105092A (en) | A kind of human body tumble recognition methods based on dynamic time warping | |
| CN108111538A (en) | Smart projector speech control system and its method based on sound groove recognition technology in e | |
| JP2017062349A (en) | Detection device and control method for the same, and computer program | |
| CN105303794A (en) | Comprehensive disposal management system for alarming by means of voiceprint recognition | |
| CN105812721A (en) | Tracking monitoring method and tracking monitoring device | |
| CN109994129A (en) | Speech processing system, method and apparatus | |
| CN110689896A (en) | Retrospective voice recognition system | |
| Rougui et al. | Audio sound event identification for distress situations and context awareness | |
| CN104243894A (en) | Audio and video fused monitoring method | |
| WO2018231133A1 (en) | Methods and devices for obtaining an event designation based on audio data | |
| CN107644651A (en) | Circuit and method for speech recognition | |
| CN203148562U (en) | Transformer substation noise imaging monitoring device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130710 |