CN112927720B - Audio anomaly detection method and device - Google Patents
Audio anomaly detection method and device Download PDFInfo
- Publication number
- CN112927720B CN112927720B CN202110112528.8A CN202110112528A CN112927720B CN 112927720 B CN112927720 B CN 112927720B CN 202110112528 A CN202110112528 A CN 202110112528A CN 112927720 B CN112927720 B CN 112927720B
- Authority
- CN
- China
- Prior art keywords
- audio signal
- time period
- preset time
- audio
- sampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 238000005070 sampling Methods 0.000 claims abstract description 246
- 230000005236 sound signal Effects 0.000 claims abstract description 208
- 230000002159 abnormal effect Effects 0.000 claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000005856 abnormality Effects 0.000 claims description 42
- 230000002596 correlated effect Effects 0.000 claims description 3
- 230000002547 anomalous effect Effects 0.000 claims 3
- 238000010586 diagram Methods 0.000 description 10
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 238000013102 re-test Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 210000005069 ears Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/78—Detection of presence or absence of voice signals
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The application discloses a method and a device for detecting audio anomalies, which belong to the technical field of audio detection and comprise the following steps: continuously sampling the audio signal output by the vehicle-mounted information entertainment navigation system according to a preset sampling frequency to obtain relevant parameters at one or more sampling points; judging whether the waveform of the output audio signal in a preset time period is distorted or not; if the waveform of the output audio signal in the preset time period is not distorted, judging whether the audio signal at one or more sampling points in the preset time period is abnormal or not according to the related parameters; and for the sampling points in the one or more sampling points, if the audio signal of the sampling point at the target moment in the preset time period is abnormal, judging the sampling point as an abnormal audio sampling point, and recording the target moment. The detection method of the audio anomaly can automatically detect, identify and record sporadic audio anomalies, improves the detection accuracy and efficiency, and reduces the detection cost.
Description
Technical Field
The application belongs to the technical field of audio detection, and particularly relates to a method and a device for detecting audio anomalies.
Background
In a vehicle-mounted entertainment system, video entertainment is the module which is most closely interacted with a user in the whole product, and is also a serious difficulty of product testing work. In the testing of car audio-visual system, the sporadic abnormal output that produces because of the audio frequency is unstable often can appear, in order to in time, accurate discovery and repair this kind of sporadic audio frequency abnormal problem, at present mainly wait for the problem to duplicate, judge audio frequency abnormal cause and record the unusual position of appearing through manual monitoring audio test system, manual monitoring judges that the precision of record problem is not high, efficiency is lower, and can produce great cost of labor.
Disclosure of Invention
In order to solve the problems in the prior art, the embodiment of the application provides a method and a device for detecting audio anomalies. The technical scheme is as follows:
the application provides a method for detecting audio anomalies, which comprises the following steps:
for an audio signal output by a vehicle-mounted information entertainment navigation system, continuously sampling the output audio signal at one or more sampling points according to a preset sampling frequency, and obtaining relevant parameters at the one or more sampling points;
judging whether the waveform of the output audio signal is distorted in a preset time period;
if the waveform of the output audio signal in the preset time period is not distorted, judging whether the audio signal at one or more sampling points in the preset time period is abnormal or not according to the related parameters;
and for the sampling point in the one or more sampling points, if the audio signal of the sampling point at the target moment in the preset time period is abnormal, judging the sampling point as an abnormal audio sampling point, and recording the target moment.
In some embodiments, the determining whether the waveform of the output audio signal is distorted within a preset time period includes:
judging whether the waveform of the output audio signal in the preset time period is distorted or not according to the total harmonic quantity of all sampling points of the output audio signal in the preset time period.
In some embodiments, the determining whether the waveform of the output audio signal in the preset time period is distorted according to the total harmonic amount of all sampling points of the output audio signal in the preset time period includes:
and if the total harmonic quantity of all sampling points in the preset time period is larger than the total harmonic quantity allowed by the input audio signal in the preset time period, judging that the waveform of the output audio signal in the preset time period is distorted.
In some embodiments, the total amount of harmonics allowed by the input audio signal over the preset time period is positively correlated with the frequency of the input audio signal.
In some embodiments, the relevant parameters at the one or more sampling points include amplitude and power of the audio signal at the one or more sampling points, and the determining whether the audio signal at the one or more sampling points is abnormal in the preset period according to the relevant parameters includes:
and judging whether the audio signals at the one or more sampling points are abnormal or not in the preset time period according to the amplitude and the power of the audio signals at the one or more sampling points.
In some embodiments, the determining whether the audio signal at the one or more sampling points is abnormal in the preset time period according to the amplitude and the power of the audio signal at the one or more sampling points includes:
and for a sampling point in the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point in the preset time period are both 0, judging that the audio signal at the sampling point is abnormal, and determining that the abnormal type of the audio signal at the sampling point is audio silence.
In some embodiments, the determining whether the audio signal at the one or more sampling points is abnormal in the preset time period according to the amplitude and the power of the audio signal at the one or more sampling points further includes:
for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point within the preset time period are not 0
Calculating an amplitude threshold and a power threshold of the audio signal at the sampling point;
if the amplitude value at the sampling point is larger than the amplitude threshold value and the power at the sampling point is larger than the power threshold value, the abnormality of the audio signal at the sampling point is judged, and the abnormality type of the audio signal at the sampling point is determined to be audio impulse noise.
In some embodiments, the calculating the amplitude threshold and the power threshold of the audio signal at the sampling point comprises:
calculating average amplitude values and average power of all sampling points of the output audio signal in the preset time period;
acquiring a user-formulated noise coefficient;
and multiplying the user-drawn noise coefficient by the average amplitude and the average power to obtain calculation results as the amplitude threshold and the power threshold.
In some embodiments, the method further comprises:
if the waveform of the output audio signal in the preset time period is distorted, storing the audio signal in the preset time period, and discarding relevant parameters of the one or more sampling points in the preset time period.
The application also provides a device for detecting the audio abnormality, which comprises:
the sampling module is used for continuously sampling the output audio signal at one or more sampling points according to a preset sampling frequency, and acquiring relevant parameters at the one or more sampling points;
the waveform distortion judging module is used for judging whether the waveform of the output audio signal in a preset time period is distorted or not;
the audio abnormality judging module is used for judging whether the audio signals at one or more sampling points in the preset time period are abnormal according to the related parameters if the waveform of the output audio signals in the preset time period is not distorted, and judging the sampling point as an audio abnormality sampling point if the audio signals at the target moment in the preset time period are abnormal;
the storage module is used for storing the audio signal with waveform distortion in the preset time period and recording the target time of the audio abnormal sampling point in the preset time period.
The technical scheme provided by the embodiment of the application has the beneficial effects that:
(1) The application provides a detection method and a detection device capable of automatically detecting, identifying and recording sporadic audio anomalies (especially audio impulse noise and silent anomalies), which can detect the audio anomalies in real time when the audio is played, and can record the played audio for later comparison, detection, retest and repair, thereby replacing the existing method for monitoring and judging by artificial squatting, realizing the automation of audio anomaly detection and reducing the labor cost.
(2) Meanwhile, the type and the occurrence time of the audio abnormality are judged by using the robot instead of the human ear, the type, the related parameters and the position of the audio abnormality are displayed and recorded, and the detection efficiency and the detection precision are improved.
(3) The waveform distortion judgment is incorporated into the audio anomaly detection method, and the sporadic audio anomalies (particularly audio impulse noise and silent anomalies) are detected on the premise of no waveform distortion, so that the accuracy and the efficiency of detecting the sporadic audio anomalies are improved.
Drawings
The above features and advantages of the present application will be better understood after reading the detailed description of the disclosed embodiments of the application in conjunction with the following drawings. It is evident that the drawings in the following description are only some embodiments of the present application and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
Fig. 1 is a schematic diagram of a method for detecting audio anomalies provided by the present application;
FIG. 2 is a flowchart illustrating an audio anomaly detection method according to the present application;
FIG. 3 is a schematic diagram of a waveform distortion detection method according to the present application;
FIG. 4 is a flowchart of a waveform distortion detection method according to the present application;
fig. 5 is a schematic diagram of an audio silence detection method according to the present application;
fig. 6 is a flowchart of an audio silence detection method provided by the present application;
fig. 7 is a schematic diagram of an audio impulse noise detection method according to the present application;
FIG. 8 is a flow chart of an audio impulse noise detection method provided by the application;
fig. 9 is a block diagram of an audio anomaly detection apparatus according to the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that the description of the present application in terms of "left", "right", lower, etc. are defined based on the relationship of orientation or position shown in the drawings, and are only for convenience of describing the present application and simplifying the description, and do not indicate or imply that the apparatus must be constructed and operated in a specific orientation, and thus should not be construed as limiting the present application. The terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. In the description of the present application, the meaning of "a plurality of" or "a number" is two or more, unless explicitly defined otherwise.
In the description of the present application, it should be noted that, unless explicitly specified and limited otherwise, such mechanical terms as "mounted," "disposed," and the like are to be construed broadly and may be fixedly connected, detachably connected, or integrally connected, for example; the device can be mechanically connected, electrically connected and communicated; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art.
The application discloses a method and a device for detecting audio anomalies, which are used for acquiring relevant parameters of sampling points by continuously sampling output audio signals, detecting whether the audio anomalies exist in a preset time period under the condition that the waveform is not distorted in the preset time period, and simultaneously storing one or more audio anomaly sampling points in the preset time period for later comparison, detection, retesting and repair. The detection method of the audio anomaly can automatically detect, identify and record sporadic audio anomalies, improves the detection accuracy and efficiency, and reduces the detection cost.
The method for detecting the audio abnormality provided by the embodiment of the application comprises the following steps:
s1: for an audio signal output by the vehicle-mounted information entertainment navigation system, continuously sampling the output audio signal at one or more sampling points according to a preset sampling frequency to obtain related parameters at one or more sampling points;
s2: judging whether the waveform of the output audio signal in a preset time period is distorted or not;
s3: if the waveform of the output audio signal in the preset time period is not distorted, judging whether the audio signal at one or more sampling points in the preset time period is abnormal or not according to the related parameters;
s4: and for a sampling point in the one or more sampling points, if the audio signal of the sampling point at the target moment in the preset time period is abnormal, judging the sampling point as an abnormal audio sampling point, and recording the target moment.
The following will describe the embodiments of the present application in detail with reference to fig. 1 to 9, and it should be noted that in the embodiments shown in fig. 1 to 9, the same or corresponding contents may be referred to each other, and will not be described in detail.
Example 1
The method for detecting audio anomalies provided by the embodiment of the application is described in detail below with reference to the embodiment and the accompanying drawings.
Fig. 1 and 2 show schematic diagrams of the audio anomaly detection method and the overall flow of the embodiment of the present application, and the following is a detailed description of the method steps of the embodiment with reference to fig. 1 and 2.
S1: and continuously sampling the output audio signal at one or more sampling points according to a preset sampling frequency for the audio signal output by the vehicle-mounted information entertainment navigation system, and obtaining relevant parameters at one or more sampling points.
Specifically, the preset sampling frequency may be set manually, and according to the detection experience, the sampling frequency set here is in positive correlation with the frequency of the input audio signal to be detected, where the frequency of the input audio signal to be detected is provided by the user, and more specifically, the preset sampling frequency is generally 5 to 10 times the frequency of the input audio signal. If the sampling frequency is 10 times of the frequency of the input audio signal, that is, if the period of the input audio signal is 1s, sampling is performed every 0.1s, the sampling points acquired at the sampling interval are most suitable, so that the characteristics of the input audio signal to be detected can be fully evaluated, and missing detection and false detection caused by too few sampling points can be avoided.
Further, the collected output audio signals may be saved to a local storage unit. Assuming that a 1000s audio signal is detected and the frequency of the audio signal is set to aHZ, if the sampling frequency is set to be 10 times of the frequency of the input audio signal, namely, the sampling frequency is set to be 10 aHZ, the detection is sampled to 1000a sampling points in total. In this embodiment, the 1000a sampling points can be monitored in real time and whether or not and where the audio signal has an abnormality can be judged, and the 1000a sampling points can be stored in a local space for later comparison, detection, retest and repair, so that the detection result is more accurate.
S2: and judging whether the waveform of the output audio signal in the preset time period is distorted or not.
Preferably, whether the waveform of the output audio signal in the preset time period is distorted is judged according to the total harmonic quantity of all sampling points of the output audio signal in the preset time period.
Preferably, if the total harmonic of all sampling points in the preset time period is greater than the total harmonic allowed in the preset time period, determining that the waveform of the output audio signal in the preset time period is distorted.
Specifically, assuming that the frequency of the audio signal is ahz, the audio signal period is 1/a s, the sampling frequency is 10 ahz, and the sampling period is 1/10a s (the sampling frequency is 10 times of the frequency of the input audio signal), further assuming that the preset time period is 1 unit audio signal period, that is, the preset time period here is 1/a s, 10 sampling points will be acquired in the preset time period, and whether the waveform of the output audio signal in the preset time period is distorted is determined, that is, whether the waveform distortion exists in the output audio signal in 1/a s is determined according to the acquired 10 sampling points. Further, if the actual total amount of harmonics of the waveform simulated from the 10 sampling points is greater than the total amount of harmonics allowed for the user-supplied input audio signal in the 1/a s period in the 1/a s period, it is determined that the waveform of the output audio signal is distorted in the preset period.
Further, the actual harmonic total XF of all sampling points within the preset time period i ＝F i -F z
Wherein F is z F for the frequency of the input audio signal provided by the user i And calculating the obtained frequency values of the audio signals of all sampling points in the preset time period through fast Fourier transform.
Preferably, the total amount of harmonics XF allowed in the preset time period when the input audio signal is output z With the frequency F of the input audio signal z Positive correlation is established.
Specifically, the smaller the value of the total harmonic of the output audio signal, the purer the tone of the output audio and the higher the quality of the product, when the total harmonic of the output audio signal is less than 1% of the frequency of the input audio signal, the total harmonic is generally indistinguishable by human ears, and when the total harmonic exceeds 10% of the frequency of the input audio signal, the human ears can clearly hear the audio distortion. Comprehensively considering the factors such as user demand, output audio quality, detection and repair cost, and the like, the total amount of permitted harmonic XF in a preset time period when the input audio signal is output z ＝F z * m, where m is E [5,10 ]]，F z Is the frequency of the input audio signal provided by the user.
Preferably, if the waveform of the output audio signal in the preset time period is distorted, the audio signal in the preset time period is saved, and the relevant parameters of one or more sampling points in the preset time period are discarded.
S3: if the waveform of the output audio signal in the preset time period is not distorted, judging whether the audio signal at one or more sampling points in the preset time period is abnormal or not according to the related parameters.
Preferably, it is determined whether the audio signal at one or more sampling points is abnormal within a preset period of time according to the amplitude and the power of the audio signal at one or more sampling points.
S4: and for the sampling points in the one or more sampling points, if the audio signal of the sampling point at the target moment in the preset time period is abnormal, judging the sampling point as an abnormal audio sampling point, and recording the target moment.
Preferably, for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point are both 0 within the preset time period, determining that the audio signal at the sampling point is abnormal, and determining that the type of abnormality of the audio signal at the sampling point is audio silence.
Preferably, for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point are not 0 within the preset time period
Calculating an amplitude threshold and a power threshold of the audio signal at the sampling point;
if the amplitude value at the sampling point is larger than the amplitude threshold value and the power at the sampling point is larger than the power threshold value, the abnormality of the audio signal at the sampling point is judged, and the abnormality type of the audio signal at the sampling point is determined to be audio impulse noise.
Further, the amplitude threshold and the power threshold of the audio signal at the sampling point include:
calculating average amplitude and average power of all sampling points of the output audio signal in a preset time period;
acquiring a user-formulated noise coefficient;
and multiplying the user-formulated noise coefficient by the average amplitude and the average power respectively, thereby obtaining an amplitude threshold value and a power threshold value.
Specifically, if the waveform of the audio signal within the preset time period is not distorted, then judging whether the audio abnormality exists within the preset time period, wherein the audio abnormality is mainly audio silence and audio impulse noise. The frequency of the audio signal is assumed as described aboveAnd if the rate is aHZ and the preset time period is 1/a s, 10 sampling points are sampled in the 1/a s, and whether the audio abnormality exists in the 1/a s is judged on the premise that the waveform in the 1/a s is not distorted. For discussion convenience, the actual amplitude of the ith sample point will be denoted as P i The actual power is recorded as W i ，1≤i≤10。
Reading the actual amplitude P of the 1 st sampling point 1 And actual power W 1 If P 1 =0 and W 1 And (0), judging the 1 st sampling point as an audio abnormal sampling point, determining the abnormal type as audio silence, and storing the sampling point with audio silence abnormality. And similarly, judging the 2 th to 10 th sampling points one by one.
Specifically, the amplitude threshold T of the 10 sampling points is calculated p And a power threshold T w ；
For the ith sample point, if P i Not equal to 0 and W i Not equal to 0, compare P i And T is p 、W i And T is w If P i ＞T p And W is i ＞T w And judging the ith sampling point as an audio abnormal sampling point, determining the abnormal type as impulse noise, and storing the sampling point with the audio impulse noise abnormality.
More specifically, for the amplitude threshold T described above p And a power threshold T w The method can be obtained by the following steps:
calculating the average amplitude of the 10 sampling pointsAnd average power +.>
Acquiring a user-formulated noise coefficient alpha, wherein the user-formulated noise coefficient is provided by the user and depends on the actual service requirement of the user;
and then calculate the amplitude thresholdPower threshold->
Further, after determining the abnormality type of the abnormal audio sampling point, the target moment in the preset time period where the sampling point is located is recorded, so that in the embodiment, not only can the audio abnormality be detected in real time when the audio is played, but also the played audio can be recorded for later comparison, detection and retest, the existing method of monitoring and judging by the artificial squatting is replaced, thereby realizing the automation of audio abnormality detection, not only reducing the labor cost, but also accurately recording the occurrence position of the audio abnormality and the related parameters thereof, ensuring that the detection result is more accurate, and providing a sufficient and comprehensive sample for later analysis and repair of the audio abnormality.
The audio anomaly detection method provided in this embodiment is further described below with reference to three specific application scenarios.
Application scenario one:
and detecting waveform distortion, audio silence and audio impulse noise of an audio signal output by the vehicle-mounted information entertainment navigation system in real time.
Specifically, according to the amplitude and the power of the audio signal at one or more sampling points, whether the audio abnormality exists at the one or more sampling points or not is judged in real time, the audio abnormality comprises audio silence and audio impulse noise, and if the audio abnormality exists, the target time of the audio abnormality sampling point in a preset time period is saved.
Judging whether the waveform of the output audio signal in the preset time period is distorted or not in real time by outputting the total harmonic quantity of all sampling points of the audio signal in the preset time period, if the waveform in the preset time period is distorted, storing the audio signal in the preset time period, further indicating that the detected audio silence and impulse noise result in the preset time period is invalid, ignoring the detected audio silence and impulse noise sampling points in the preset time period, and effectively detecting the audio silence and impulse noise result in the waveform distortion detection result in the waveform non-distortion time period.
And (2) an application scene II:
and detecting audio silence and audio impulse noise of an audio signal output by the vehicle-mounted information entertainment navigation system in real time, continuously sampling in real time, acquiring the amplitude and frequency of the audio signal at one or more sampling points, and detecting waveform distortion of the sampled audio signal.
Specifically, according to the amplitude and the power of the audio signal at one or more sampling points, whether the audio abnormality exists at the one or more sampling points or not is judged in real time, the audio abnormality comprises audio silence and audio impulse noise, and if the audio abnormality exists, the target time of the audio abnormality sampling point in a preset time period is saved.
Continuously sampling and storing the sampled audio signals in real time, and simultaneously acquiring the amplitude and the frequency of the audio signals at one or more sampling points; judging whether waveform distortion exists in the stored audio signal, more specifically judging whether the waveform of the output audio signal in the preset time period is distorted or not according to the total harmonic quantity of all sampling points in the preset time period, if the waveform in the preset time period is distorted, the detected audio silence and impulse noise result in the preset time period is invalid, the detected audio silence and impulse noise sampling points in the preset time period are ignored, and the detected audio silence and impulse noise result in the waveform distortion detection result is valid in the waveform undistorted time period.
And (3) an application scene III:
and continuously sampling the audio signals output by the vehicle-mounted information entertainment navigation system in real time, acquiring relevant parameters such as amplitude, frequency, power and the like of the audio signals at one or more sampling points, and detecting waveform distortion, audio silence and audio impulse noise of the sampled audio signals.
Specifically, the sampled audio signals are continuously sampled and stored in real time, and related parameters such as frequency, amplitude, power and the like of the audio signals at one or more sampling points are obtained. Waveform distortion, audio silence, audio impulse noise detection are performed on the stored audio signal. More specifically, according to the amplitude and the power of the audio signal at one or more sampling points, judging whether audio anomalies exist at one or more sampling points in the sampled audio signal, wherein the audio anomalies comprise audio silence and audio impulse noise, and if the audio anomalies exist, storing target moments of the audio anomaly sampling points in a preset time period; judging whether the waveform of the output audio signal in the preset time period is distorted or not through the total harmonic quantity of all sampling points in the preset time period, if the waveform in the preset time period is distorted, storing the audio signal in the preset time period, further indicating that the detected audio silence and impulse noise result in the preset time period is invalid, ignoring the detected audio silence and impulse noise sampling points in the preset time period, and indicating that the detected audio silence and impulse noise result in the waveform undistorted time period in the waveform distortion detection result is valid.
Example 2
The waveform distortion detection method provided by the embodiment of the present application will be described in detail with reference to the embodiment and the accompanying drawings.
Fig. 3 and 4 are schematic diagrams of the waveform distortion detection method of the present application and flowcharts of the embodiments, and the following is a detailed description of the method steps of the present embodiment with reference to fig. 3 and 4.
Judging whether the waveform of the output audio signal in the preset time period is distorted or not according to the total harmonic quantity of all sampling points of the output audio signal in the preset time period:
s21: calculating the actual harmonic total XF of all sampling points in the preset time period i ；
S22: calculating the total amount of permitted harmonics XF of the input audio signal within the preset time period z ；
Wherein XF i ＝F i -F z ，
F z F for the frequency of the input audio signal provided by the user i Calculating frequency values of all sampling point audio signals in the preset time period through fast Fourier transform;
XF z ＝F z *m％，
m∈[5,10]，F z a frequency of an input audio signal provided by a user;
s23: if XF i ＞XF z The waveform of the output audio signal is distorted for the preset period of time.
S24: the audio signal of the preset time period is saved, and relevant parameters of one or more sampling points in the preset time period are discarded.
Example 3
The following describes only the audio silence detection method according to the embodiment of the present application in detail with reference to the embodiment and the accompanying drawings.
Fig. 5 and 6 are schematic diagrams of the audio silence detection method of the present application and a flowchart of an embodiment, and the following is a detailed description of the method steps of the present embodiment with reference to fig. 5 and 6.
The output audio signal is under the precondition that the waveform in the preset time period is not distorted:
s31: reading the amplitude and the power of the audio signal at one or more sampling points in the preset time period;
s321: for a sampling point in the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point in the preset time period are both 0, determining that the audio signal at the sampling point is abnormal and determining that the abnormal type of the audio signal at the sampling point is audio silence;
s33: and saving the audio abnormal sampling points.
Example 4
The following describes only the audio impulse noise detection method provided by the embodiment of the present application in detail with reference to the embodiment and the accompanying drawings.
Fig. 7 and 8 are schematic diagrams of the audio impulse noise detection method of the present application and flowcharts of embodiments, and the following is a detailed description of the method steps of the present embodiment with reference to fig. 7 and 8.
The output audio signal is under the precondition that the waveform in the preset time period is not distorted:
s31: reading the amplitude and the power of the audio signal at one or more sampling points in the preset time period;
s322: for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point are not 0 within the preset time period
Calculating an amplitude threshold T of the audio signal at the sampling point p And a power threshold T w ，
Wherein the amplitude thresholdPower threshold->
Alpha is the noise coefficient formulated for the user,and->The average amplitude and average power of one or more sampling points in a preset time period are respectively;
for the ith sample point, if P i ＞T p And W is i ＞T w Determining that the audio signal at the sampling point is abnormal, and determining that the abnormal type of the audio signal at the sampling point is audio impulse noise;
s33: and saving the audio abnormal sampling points.
Example 5
The application also provides a device for detecting the audio abnormality, and the device for detecting the audio abnormality provided by the embodiment of the application is only described in detail below with reference to the embodiment and the attached drawings.
Fig. 9 is a block diagram showing an embodiment of the audio abnormality detection apparatus of the present application, and the following is a detailed description of the apparatus structure of the present embodiment, with reference to fig. 9.
The application provides a detection device for audio anomalies, which comprises:
a sampling module 10, configured to continuously sample the output audio signal at one or more sampling points according to a preset sampling frequency, and obtain relevant parameters at the one or more sampling points;
a waveform distortion judging module 20 for judging whether the waveform of the output audio signal is distorted in a preset period of time;
the audio anomaly determination module 30 is configured to determine whether the audio signal at one or more sampling points in the preset time period is abnormal according to the related parameters if the waveform of the output audio signal in the preset time period is not distorted, and determine the sampling point as an audio anomaly sampling point if the audio signal at the target time in the preset time period is abnormal;
and the storage module 10 is used for storing the audio signal with waveform distortion in a preset time period and recording the target time of the audio abnormal sampling point in the preset time period.
Preferably, the waveform distortion judging module 20 is specifically configured to judge whether the waveform of the output audio signal in the preset time period is distorted according to the total harmonic amounts of all sampling points of the output audio signal in the preset time period.
Preferably, the waveform distortion determining module 20 is specifically configured to determine that the waveform of the output audio signal is distorted in the preset time period if the total harmonic of all sampling points in the preset time period is greater than the total harmonic allowed by the input audio signal in the preset time period.
Further, the total amount of harmonics allowed in the input audio signal within the preset time period is positively correlated with the frequency of the input audio signal.
Preferably, the audio anomaly determination module 30 is specifically configured to determine whether the audio signal at one or more sampling points is abnormal in a preset period according to the amplitude and the power of the audio signal at one or more sampling points.
Preferably, the audio anomaly determination module 30 is specifically configured to determine, for a sampling point of the one or more sampling points, that the audio signal at the sampling point is abnormal if the amplitude and the power of the audio signal at the sampling point are both 0 within a preset period of time, and determine that the type of the abnormality of the audio signal at the sampling point is audio silence.
Preferably, the audio anomaly determination module 30 is specifically configured to, for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point are not 0 within the preset time period
Calculating an amplitude threshold and a power threshold of the audio signal at the sampling point;
if the amplitude value at the sampling point is larger than the amplitude threshold value and the power at the sampling point is larger than the power threshold value, the abnormality of the audio signal at the sampling point is judged, and the abnormality type of the audio signal at the sampling point is determined to be audio impulse noise.
Further, the steps of calculating the amplitude threshold and the power threshold are as follows:
calculating average amplitude values and average power of all sampling points of the output audio signal in the preset time period;
acquiring a user-formulated noise coefficient;
and multiplying the user-drawn noise coefficient by the average amplitude and the average power respectively to obtain the amplitude threshold and the power threshold.
The technical scheme provided by the embodiment of the application has the beneficial effects that:
(1) The application provides a detection method and a detection device capable of automatically detecting, identifying and recording sporadic audio anomalies (especially audio impulse noise and silent anomalies), which can detect the audio anomalies in real time when the audio is played, and can record the played audio for later comparison, detection, retest and repair, thereby replacing the existing method for monitoring and judging by artificial squatting, realizing the automation of audio anomaly detection and reducing the labor cost.
(2) Meanwhile, the type and the occurrence time of the audio abnormality are judged by using the robot instead of the human ear, the type, the related parameters and the position of the audio abnormality are displayed and recorded, and the detection efficiency and the detection precision are improved.
(3) The waveform distortion judgment is incorporated into the audio anomaly detection method, and the sporadic audio anomalies (particularly audio impulse noise and silent anomalies) are detected on the premise of no waveform distortion, so that the accuracy and the efficiency of detecting the sporadic audio anomalies are improved.
Any combination of the above optional solutions may be adopted to form an optional embodiment of the present application, which is not described herein.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
The foregoing description of the preferred embodiments of the application is not intended to limit the application to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the application are intended to be included within the scope of the application.
What is not described in detail in this specification is prior art known to those skilled in the art.
Claims (9)
1. A method for detecting an audio anomaly, comprising:
for an audio signal output by a vehicle-mounted information entertainment navigation system, continuously sampling the output audio signal at one or more sampling points according to a preset sampling frequency, and obtaining relevant parameters at the one or more sampling points;
judging whether the waveform of the output audio signal in a preset time period is distorted or not according to the total harmonic quantity of all sampling points of the output audio signal in the preset time period;
if the waveform of the output audio signal in the preset time period is not distorted, judging whether the audio signal at one or more sampling points in the preset time period is abnormal or not according to the related parameters;
and for the sampling point in the one or more sampling points, if the audio signal of the sampling point at the target moment in the preset time period is abnormal, judging the sampling point as an abnormal audio sampling point, and recording the target moment.
2. The method for detecting an audio anomaly according to claim 1, wherein the determining whether the waveform of the output audio signal in a preset time period is distorted according to the total harmonic amounts of all sampling points of the output audio signal in the preset time period comprises:
and if the total harmonic quantity of all sampling points in the preset time period is larger than the total harmonic quantity allowed by the input audio signal in the preset time period, judging that the waveform of the output audio signal in the preset time period is distorted.
3. The method of claim 2, wherein the total amount of harmonics allowed in the input audio signal over the predetermined period of time is positively correlated with the frequency of the input audio signal.
4. A method of detecting an audio anomaly according to any one of claims 1 to 3 wherein the associated parameters at the one or more sample points include the amplitude and power of the audio signal at the one or more sample points, and wherein determining whether the audio signal at the one or more sample points is anomalous for the predetermined period of time based on the associated parameters comprises:
and judging whether the audio signals at the one or more sampling points are abnormal or not in the preset time period according to the amplitude and the power of the audio signals at the one or more sampling points.
5. The method for detecting an audio anomaly according to claim 4, wherein the determining whether the audio signal at the one or more sampling points is anomalous within the preset time period according to the amplitude and the power of the audio signal at the one or more sampling points comprises:
and for a sampling point in the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point in the preset time period are both 0, judging that the audio signal at the sampling point is abnormal, and determining that the abnormal type of the audio signal at the sampling point is audio silence.
6. The method for detecting an audio anomaly according to claim 5, wherein the determining whether the audio signal at the one or more sampling points is anomalous within the preset time period according to the amplitude and the power of the audio signal at the one or more sampling points further comprises:
for a sampling point of the one or more sampling points, if the amplitude and the power of the audio signal at the sampling point within the preset time period are not 0
Calculating an amplitude threshold and a power threshold of the audio signal at the sampling point;
if the amplitude value at the sampling point is larger than the amplitude threshold value and the power at the sampling point is larger than the power threshold value, the abnormality of the audio signal at the sampling point is judged, and the abnormality type of the audio signal at the sampling point is determined to be audio impulse noise.
7. The method of claim 6, wherein calculating the amplitude threshold and the power threshold of the audio signal at the sampling point comprises:
calculating average amplitude values and average power of all sampling points of the output audio signal in the preset time period;
acquiring a user-formulated noise coefficient;
and multiplying the user-drawn noise coefficient by the average amplitude and the average power to obtain calculation results as the amplitude threshold and the power threshold.
8. A method of detecting an audio anomaly as claimed in any one of claims 1 to 3, further comprising:
if the waveform of the output audio signal in the preset time period is distorted, storing the audio signal in the preset time period, and discarding relevant parameters of the one or more sampling points in the preset time period.
9. An audio anomaly detection apparatus, comprising:
the sampling module is used for continuously sampling the output audio signals at one or more sampling points according to a preset sampling frequency, and acquiring relevant parameters at the one or more sampling points;
the waveform distortion judging module is used for judging whether the waveform of the output audio signal in the preset time period is distorted or not according to the total harmonic quantity of all sampling points of the output audio signal in the preset time period;
the audio abnormality judging module is used for judging whether the audio signals at one or more sampling points in the preset time period are abnormal according to the related parameters if the waveform of the output audio signals in the preset time period is not distorted, and judging the sampling point as an audio abnormality sampling point if the audio signals at the target moment in the preset time period are abnormal;
the storage module is used for storing the audio signal with waveform distortion in the preset time period and recording the target time of the audio abnormal sampling point in the preset time period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110112528.8A CN112927720B (en) | 2021-01-27 | 2021-01-27 | Audio anomaly detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110112528.8A CN112927720B (en) | 2021-01-27 | 2021-01-27 | Audio anomaly detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112927720A CN112927720A (en) | 2021-06-08 |
| CN112927720B true CN112927720B (en) | 2023-09-29 |
Family
ID=76167145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110112528.8A Active CN112927720B (en) | 2021-01-27 | 2021-01-27 | Audio anomaly detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112927720B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113466666A (en) * | 2021-07-06 | 2021-10-01 | 格力电器（郑州）有限公司 | Mainboard detection device and detection method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103855A (en) * | 2009-12-16 | 2011-06-22 | 北京中星微电子有限公司 | Method and device for detecting audio clip |
| JP2012085040A (en) * | 2010-10-08 | 2012-04-26 | Mitsubishi Electric Corp | On-vehicle audio device |
| CN103033663A (en) * | 2012-12-24 | 2013-04-10 | 电子科技大学 | Anomaly detection method for three-dimensional waveform data |
| CN107086039A (en) * | 2017-05-25 | 2017-08-22 | 北京小鱼在家科技有限公司 | A kind of acoustic signal processing method and device |
| CN208888339U (en) * | 2018-06-11 | 2019-05-21 | 汉能移动能源控股集团有限公司 | A kind of fault detection system of application of solar energy circuit |
| CN111918196A (en) * | 2019-05-08 | 2020-11-10 | 腾讯科技（深圳）有限公司 | Method, device and equipment for diagnosing recording abnormity of audio collector and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10879864B2 (en) * | 2005-09-27 | 2020-12-29 | Ronald Quan | Method and apparatus to evaluate audio equipment for dynamic distortions and or differential phase and or frequency modulation effects |
-
2021
- 2021-01-27 CN CN202110112528.8A patent/CN112927720B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103855A (en) * | 2009-12-16 | 2011-06-22 | 北京中星微电子有限公司 | Method and device for detecting audio clip |
| JP2012085040A (en) * | 2010-10-08 | 2012-04-26 | Mitsubishi Electric Corp | On-vehicle audio device |
| CN103033663A (en) * | 2012-12-24 | 2013-04-10 | 电子科技大学 | Anomaly detection method for three-dimensional waveform data |
| CN107086039A (en) * | 2017-05-25 | 2017-08-22 | 北京小鱼在家科技有限公司 | A kind of acoustic signal processing method and device |
| CN208888339U (en) * | 2018-06-11 | 2019-05-21 | 汉能移动能源控股集团有限公司 | A kind of fault detection system of application of solar energy circuit |
| CN111918196A (en) * | 2019-05-08 | 2020-11-10 | 腾讯科技（深圳）有限公司 | Method, device and equipment for diagnosing recording abnormity of audio collector and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 一种扬声器异常音的时域特征检测方法;韦峻峰;杨益;温周斌;冯海泓;;振动与冲击(10);第122-128页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112927720A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3163574B1 (en) | Method and apparatus for detecting abnormal frame | |
| CN108229441B (en) | Classroom teaching automatic feedback system and feedback method based on image and voice analysis | |
| KR950002442B1 (en) | Checking audio system | |
| AU773512B2 (en) | Method and system for measurement of speech distortion from samples of telephonic voice signals | |
| CN103546853A (en) | Speaker abnormal sound detecting method based on short-time Fourier transformation | |
| CN112927720B (en) | Audio anomaly detection method and device | |
| CN113259832B (en) | Microphone array detection method and device, electronic equipment and storage medium | |
| JP2008015443A (en) | Apparatus, method and program for estimating noise suppressed voice quality | |
| US20090304195A1 (en) | Method and device for diagnosing the operating state of a sound system | |
| CN110503982B (en) | Voice quality detection method and related device | |
| WO2021203654A1 (en) | Audio test system and audio test method for electronic product | |
| CN111182434A (en) | Conference system microphone anomaly detection method, conference system and test system | |
| CN113593536B (en) | Device and system for detecting voice recognition accuracy | |
| CN111355949B (en) | Method for constructing audio and video multimedia database and evaluating subjective quality of multimedia | |
| CN102208190B (en) | Measurement method and device for inhibiting unstable noise equipment convergence time | |
| CN115219955A (en) | Transformer winding deformation detection system and detection method thereof | |
| CN110099350A (en) | The test method of power amplifier | |
| CN104994377A (en) | Television machine vibration test method | |
| CN111148005B (en) | Method and device for detecting mic sequence | |
| CN113596698A (en) | Microphone reception monitoring device and method | |
| CN107172558A (en) | A kind of novel horn fault detection method | |
| WO2023272899A1 (en) | Shipborne communication monitoring system | |
| CN113015076A (en) | Mobile terminal, passenger information service system and test system thereof | |
| JPH11118593A (en) | Plant failure monitoring device | |
| WO2022262073A1 (en) | Method and system for training device model of target device by means of sample doping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |