CN103983345B - A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method - Google Patents
A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method Download PDFInfo
- Publication number
- CN103983345B CN103983345B CN201410122439.1A CN201410122439A CN103983345B CN 103983345 B CN103983345 B CN 103983345B CN 201410122439 A CN201410122439 A CN 201410122439A CN 103983345 B CN103983345 B CN 103983345B
- Authority
- CN
- China
- Prior art keywords
- frequency
- sound
- signal
- voicing
- sin
- 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.)
- Expired - Fee Related
Links
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
It is the type of subject in ambient noise that single-frequency, which has voicing signal, the present invention be directed to the single-frequency in ambient noise voicing signal automatic monitoring method, its physical characterization and the auditory properties of human ear are made to realize high accuracy fitting, this method set single-frequency has real-time collection, mathematical description and the physical characterization of voicing signal.Its core is to be directed to A-weighted sound level core algorithm in environment noise monitoring to be improved, propose the multinomial single-frequency noise modifying factor mathematical description relational expression for adding trigonometric function with logarithmic function, and realize undetermined coefficient and initial phase angle parameter optimization in the mathematical description relational expression, by being fitted experiment test in high precision for 40phon equal loudness contours, it is fitted absolute error and is less than 0.15dB, and have found identical changing rule for the equal loudness contour of different loudness.The measurement error of precision sound level meter is about native 1dB, and common sound level meter is about native 3dB.Using method, its measurement error is less than 0.15dB in the present invention, and precision fully meets requirement.Not yet find similar technique at present.
Description
Technical field
The present invention relates to a kind of single-frequency based on human hearing characteristic voicing signal automatic monitoring method.
Background technology
Usual people are to reflect tested noise sound level height by subjective assessment parameter.Sound pressure level etc. is used as acoustic environment matter
The primary subjective parameter of amount, subjective parameter corresponding to it is sound level, really by the revised sound pressure level of weighted.International standard
Change tissue ISO226：2003 and chinese national standard GB/T4963-2007 has revised equal loudness contour respectively, referring to Fig. 1.Thus may be used
See, equal loudness contour is the relation curve of sound pressure level and frequency under the conditions of description etc. rings, and is one of important aural signature.That is sound
Frequency it is different, the curve that sound pressure level changes with frequency when it and 1kHz pure tones etc. are loud is referred to as equal loudness contour.Equal loudness contour is one
Individual statistic curve, it is contemplated that the aural signature of crowd.As shown in figure 1, correspond to the acoustic pressure of different frequency in figure on every curve
Level differs, but the response that human ear is felt is the same, and a numeral is marked with every curve, is volume unit.By
Equal loudness contour race can learn：When volume is smaller, human ear is to high bass perception deficiency；And volume it is larger when, high bass perception
Fully；People is most sensitive to sound between 1kHz ~ 4kHz.For example, the sound intensity level of 1kHz sound is 60dB, and another frequency
Sound sound as 60dB 1kHz sound ring, then the loudness level of this sound is 60 phon；And 50dB
100Hz pure tones and 40dB 1kHz pure tones etc. ring, i.e., both are in the phon equal loudness contours of same 40.Around 1kHz's
In intermediate frequency range, contour of equal loudness compares relatively low, illustrates that response of the human ear to intermediate frequency is sensitive.Low frequency outside this scope
With high frequency both sides, contour of equal loudness tilts, and illustrates that human ear declines to low frequency and the sensitive of high-frequency sound, so that when frequency is less than
20Hz and during higher than 2kHz, it is necessary to which the very big sound intensity is possible to the presence of perceived sounds.
When using sound level meter measurement noise, what sound pressure sensor collected is sound pressure signal.If it is not made any
Processing just exports, and what is obtained will be the linear sound level unrelated with frequency.According to the physilogical characteristics of human ear, the sense of hearing of people is not only depended on
In the sound intensity but also relevant with frequency, i.e. perhaps the sensation for the sound that the acoustical signal people of identical acoustic pressure different frequency is heard can
There is different, it is desirable to the physiological property that the noise sound of apparatus measures can meet human ear.In view of human ear to difference
The acoustical signal of frequency, which is listened, to be distinguished and filtering characteristic, therefore, processing is filtered with reference to equal loudness contour, the frequency content sensitive to human ear
It is reinforced, and appropriate decay is then carried out to the insensitive frequency content of human ear, is use up in the hope of the subjective feeling with human auditory system
May be consistent.The method of this amendment is referred to as spectrum overlapping, and the sound level measured by weighting network is referred to as weighted sound level.It there is now
A, a variety of weighting networks such as B, C, D.Equal loudness contour represents subjective sensation mistake of the people to sound with weighting network with loudness level
In complexity, for simplicity, selected three curves in equal loudness contour, one be 40phon curve, represent low sound pressure levels
The sensation of loudness；One be 70 phon curve, represent the sensation of loudness of moderate strength；One be 100 phon curve, generation
Sensation of loudness during table high sound intensity.Shape according to this three curves devises tri- weighting networks of A, B, C.A weighting networks are special
Linearity curve corresponds to inverted 40 phon equal loudness contours, and B weighting networks curve corresponds to inverted 70phon equal loudness contours, C meters
Power network curve corresponds to inverted 100 phon equal loudness contours.ISO recommend standard in, noise measuring method has been made with
Lower regulation：1. when linear sound level is not less than i.e. Lin≤60dB, using the weighting network of A characteristic curve；2. work as 60dB<Lin<
During l20dB, using the weighting network of B characteristic curve；3. as Lin >=120dB, then the characteristic weighting networks of C must be used.
It was verified that no matter noise intensity is high or low, A sound levels can reflect people to noise loudness and noisy sensation, mesh well
Before, using A sound levels and C sound levels as evaluation criterion.Although A sound levels are as international standard, it reflects human ear to a certain extent
Auditory properties, but larger deviation can be embodied at some frequencies.There is voicing signal to make it to accurately characterize single-frequency
Matched with human hearing characteristic, and set up distributed environment automatic monitoring system for noise to need high precision monitor node,
And for monitoring node functional requirement：Single-point ambient noise signal can either be detected, analyzed, handled and shown, may be used also
So that significant data is passed through into the long-range real-time Transmission of real-time performance.Therefore, a kind of Novel ring based on virtual instrument technique is invented
Border noise automatic monitoring method.
The content of the invention
It is the type of subject in ambient noise that single-frequency, which has voicing signal, and the present invention be directed to the single-frequency in ambient noise tune
Acoustical signal automatic monitoring method, its physical characterization and the auditory properties of human ear are made to realize high accuracy fitting, this method set list
Frequency has real-time collection, mathematical description and the physical characterization of voicing signal.Its core is to be directed to A-weighted sound level in environment noise monitoring
Core algorithm is improved, it is proposed that is closed with the multinomial single-frequency noise modifying factor mathematical description of logarithmic function plus trigonometric function
It is formula.Single-frequency is set to have the physical characterization of voicing signal and the auditory properties of human ear to realize high accuracy fitting by amendment.
Brief description of the drawings
Fig. 1 is standard equal loudness contour figure；
Fig. 2 has voicing signal automatic monitoring method flow block diagram for consistent with prior art；
Fig. 3 is that the single-frequency consistent with the specific embodiment of the invention has voicing signal automatic monitoring method flow；
Fig. 4 is the improvement A weighting network frequency characteristic correction value matched curve figure consistent with the specific embodiment of the invention；
Fig. 5 is the improvement A weighting network frequency characteristic fitting of a polynomial error curve consistent with the specific embodiment of the invention
Figure.
Specific embodiment
In the prior art, there is the FB(flow block) of voicing signal automatic monitoring method referring to Fig. 2.Signal collection in real time is substantially all
Ambient noise signal is gathered using professional sound pressure sensor, for example using electret testing capacitor microphone.Signal condition link
Mainly acoustic pressure is sensed using a series of circuits such as preamplifier, weighted amplification, LMS detections and direct current amplification, A/D conversion
Device collection ambient noise signal is sent to microprocessor after being nursed one's health and performed mathematical calculations, and calculates instantaneous sound level, equivalent sound level
Deng mathematical description parameter, then traditionally characterizing method according to formula（1）A weighting network frequency characteristic amendments, it is final to realize
The physical characterization of ambient noise signal.
And the present invention has voicing signal automatic monitoring method flow using single-frequency, idiographic flow is referring to Fig. 3.With PC computers
It is hardware with integrated sound card and sound pressure sensor, using LabVIEW as Software Development Platform.First, it is reliably general using stability
Flow-through electric capacity sound pressure sensor realizes that sound pressure signal gathers in real time by PC Automated library system sound cards；It secondly, will collect micro-
Pressure signal amplifies conditioning in LabVIEW platform while carries out spectrum analysis and fundamental frequency measure, and amplified signal is directly passed through
Computing is converted into instantaneous sound level；Then instantaneous sound level is converted into equivalent sound level by the period again, while voicing letter is had according to single-frequency
Number characteristic frequency, with formula（2）Based on carry out the amendment of weighted frequency to realize curve matching；Finally by improved A frequencies
The revised sound pressure level of rate weighting network has voicing signal to realize the physical characterization for meeting human ear characteristic to single-frequency.Said process is real
It is that its overall process is by certainly in order to realize the real-time collection of ambient noise signal, processing, display and the storage of critical data in matter
The environment noise monitoring system based on LabVIEW of row exploitation is automatically performed, and is needed before use according to the different sensitivity of use
Probe is demarcated to instrument.
Specific test simulation result is referring to Fig. 4, and two curves are not only numerically variant, for example, in the Hz of low-frequency range 200
Following fair curve is generally less than original standard A weighting networks frequency characteristic correction value, and frequency is lower, and difference is bigger, 20Hz frequencies
The corresponding nearly 10dB of sound level correction value difference, illustrate after 200 below Hz low-frequency ranges carry out A weighting network frequency characteristic amendments
Loudness than being actually hearing is eager to excel；Mid Frequency between 300Hz ~ 1kHz, two curves coincide substantially；1kHz ~ 2kHz,
Original standard A weighting network frequency characteristics correction value is higher than actual curve between this two sections of 5kHz ~ 16kHz, reflects test ratio
Actual loudness is weak；Original standard A weighting network frequency characteristics correction value is less than actual curve between 2kHz ~ 5kHz, reflects
Test is eager to excel than actual loudness, this section exactly hearing sound sensitive area.As can be seen here, A weighting network frequency characteristics are improved to repair
Positive curve is very necessary.Can the key issue that improved A weighting networks frequency characteristic is run into be to be easy to mathematical operation
And solution, whether its error of fitting, which meets, requires.Therefore, having carried out a series of exploration, seek to polynomial fitting curve
Optimized algorithm, its calculate referring to formula（2）.
The present invention is that have the voicing signal correction factor to carry out mathematical description to single-frequency by innovatory algorithm, and realizes the number
Undetermined coefficient and initial phase angle parameter optimization in description relational expression are learned, by the way that fitting experiment is surveyed in high precision for 40phon equal loudness contours
Examination, it is fitted absolute error and is less than 0.15dB（Referring to Fig. 5）, and have found identical change for the equal loudness contour of different loudness
Law.The measurement error of precision sound level meter is about native 1dB, and common sound level meter is about native 3dB.Measured using the method for the present invention
Error is less than 0.15dB, and precision fully meets requirement.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
In the case of departing from the principle and objective of the present invention a variety of change, modification, replacement and modification can be carried out to these embodiments, this
The scope of invention is limited by claim and its equivalent.
Claims (1)
1. a kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method, applied to PC computers, the PC
Computer can be connected with electric capacity sound pressure sensor, and the PC computer installations have LabVIEW platform, and it includes：
Sound pressure signal is gathered using electric capacity sound pressure sensor in real time by the integrated sound card of the PC computers；
The micro-pressure signal collected is amplified in the LabVIEW platform and nurses one's health while carry out spectrum analysis and fundamental frequency measure；
Amplified signal is directly converted into instantaneous sound level by computing；
Instantaneous sound level is converted into equivalent sound level by the period, while the characteristic frequency for having voicing signal according to single-frequency carries out weighted frequency
Rate amendment is to realize curve matching；
There is voicing signal to realize to single-frequency by the improved revised sound pressure level of A spectrum overlappings network and meet the thing of human ear characteristic
Reason characterizes；
The wherein described characteristic frequency for having voicing signal according to single-frequency carries out the amendment of weighted frequency to realize that curve matching is basis
What below equation was carried out：
Y (f)=a0+a1x+a2x2+a3x3+a4x4+a5x5+a6x6+a7x7+a8x8+a9cos(x+φ1)2+a10sin(x+θ0)
+a11sin(x+θ1)2+a12sin(x+θ2)3+a13sin(x+θ3)5+a14sin(x+θ4)7+a15sin(x+θ5)9
In formula：X=logf, f, which represent single-frequency, the frequency of voicing signal, ai(i=0,1......15) is optimized coefficients,θi
(i=0,1......5) it is initial phase angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410122439.1A CN103983345B (en) | 2014-03-28 | 2014-03-28 | A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410122439.1A CN103983345B (en) | 2014-03-28 | 2014-03-28 | A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103983345A CN103983345A (en) | 2014-08-13 |
| CN103983345B true CN103983345B (en) | 2017-11-17 |
Family
ID=51275405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410122439.1A Expired - Fee Related CN103983345B (en) | 2014-03-28 | 2014-03-28 | A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103983345B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104729677B (en) * | 2014-12-31 | 2017-10-03 | 清华大学 | A kind of time-domain digital weighted method of nonstationary noise signal |
| CN106092306B (en) * | 2016-06-02 | 2019-04-30 | 青岛歌尔声学科技有限公司 | A kind of acoustic pressure test method and acoustic pressure test macro |
| CN108871549A (en) * | 2018-03-19 | 2018-11-23 | 广州亿航智能技术有限公司 | Intelligent aircraft NVH test device, system and test method |
| CN112879278B (en) * | 2021-01-11 | 2022-09-30 | 苏州欣皓信息技术有限公司 | Pump station unit fault diagnosis method based on noise signal A weighting analysis |
| CN112954115B (en) * | 2021-03-16 | 2022-07-01 | 腾讯音乐娱乐科技（深圳）有限公司 | Volume adjusting method and device, electronic equipment and storage medium |
| CN113421539B (en) * | 2021-07-19 | 2023-10-10 | 北京安声浩朗科技有限公司 | Active noise reduction method and device, electronic equipment and computer readable storage medium |
-
2014
- 2014-03-28 CN CN201410122439.1A patent/CN103983345B/en not_active Expired - Fee Related
Non-Patent Citations (5)
| Title |
|---|
| 基于虚拟仪器技术的环境噪声监测仪的涉及;刘祥楼等;《中北大学学报（自然科学版）》;20090630;第30卷(第6期);全文 * |
| 基于阶比分析的风力发电机噪声音调判定;徐冠基等;《振动、测试与诊断》;20100831;第30卷(第4期);全文 * |
| 声级计的工作原理;钟国策;《电声技术》;20020131(第1期);第2部分 * |
| 扩展型多项式曲线拟合;李天资;《右江民族医学院学报》;19980331;第20卷;第1页左栏 * |
| 用A-计权声级评价测听环境噪声;章句才;《电声技术》;19931231(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103983345A (en) | 2014-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103983345B (en) | A kind of single-frequency based on human hearing characteristic has voicing signal automatic monitoring method | |
| CN105163262B (en) | A kind of loudspeaker sound detection method and detecting system | |
| Alamir et al. | Subjective responses to wind farm noise: A review of laboratory listening test methods | |
| CN101426168A (en) | Sounding body abnormal sound detection method and system | |
| Katalin | Studying noise measurement and analysis | |
| US11818554B2 (en) | Headset playback acoustic dosimetry | |
| CN112487865A (en) | Automatic loudspeaker classification method based on machine learning | |
| CN109982637A (en) | The method for accurately estimating pure tone threshold using the audio system not referred to | |
| Håkansson et al. | The mechanical impedance of the human skull via direct bone conduction implants | |
| Robinson et al. | Comparative study of the performance of smartphone-based sound level meter apps, with and without the application of a ½” IEC-61094-4 working standard microphone, to IEC-61672 standard metering equipment in the detection of various problematic workplace noise environments | |
| CN111128226A (en) | Device and method for detecting noise sound quality | |
| Padilla-Ortiz et al. | Smartphones, a tool for noise monitoring and noise mapping: an overview | |
| CN109933933B (en) | Noise treatment method and equipment | |
| CN112098939A (en) | Method and device for identifying and evaluating noise pollution source | |
| CN108682430A (en) | A kind of method of speech articulation in objective evaluation room | |
| Zhao et al. | Development of an automatic classifier for the prediction of hearing impairment from industrial noise exposure | |
| CN117041847B (en) | Adaptive microphone matching method and system for hearing aid | |
| CN108344503A (en) | A kind of high-speed transplanter noise qualities evaluation system | |
| CN117688291A (en) | Vibration comfort evaluation method, equipment and medium for hydropower house | |
| Mariscal-Ramirez et al. | Knowledge-based wireless sensors using sound pressure level for noise pollution monitoring | |
| CN113155271B (en) | Acoustic vibration detection method, system, terminal and medium | |
| CN108513226A (en) | A kind of method and device of detection two-channel equipment stereophonic effect | |
| Sakurai et al. | A computational software for noise measurement and toward its identification | |
| CN115813379B (en) | Middle ear sound admittance testing method and device and earphone | |
| WO2005106407A1 (en) | A method of objectively determining subjective properties of a binaural sound signal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171117 Termination date: 20200328 |