CN110033784A - A kind of detection method of audio quality, device, electronic equipment and storage medium - Google Patents
A kind of detection method of audio quality, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110033784A CN110033784A CN201910284120.1A CN201910284120A CN110033784A CN 110033784 A CN110033784 A CN 110033784A CN 201910284120 A CN201910284120 A CN 201910284120A CN 110033784 A CN110033784 A CN 110033784A
- Authority
- CN
- China
- Prior art keywords
- audio
- detected
- frequency
- region signal
- energy
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000001228 spectrum Methods 0.000 claims description 83
- 238000004891 communication Methods 0.000 claims description 25
- 238000000034 method Methods 0.000 claims description 25
- 238000000605 extraction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 235000013399 edible fruits Nutrition 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 15
- 238000004364 calculation method Methods 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000005236 sound signal Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 230000002093 peripheral effect Effects 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 230000000712 assembly Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 235000015170 shellfish Nutrition 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000003068 static 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
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/18—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being spectral information of each sub-band
-
- 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
- G10L25/60—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 for measuring the quality of voice signals
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Quality & Reliability (AREA)
- Stereophonic System (AREA)
Abstract
The embodiment of the present application provides detection method, device, electronic equipment and the storage medium of a kind of audio quality, which comprises obtains multiple audios to be detected;Extract the audio frequency characteristics of each audio to be detected, wherein audio frequency characteristics are the feature for identifying the audio quality of the audio to be detected;According to the audio frequency characteristics of each audio to be detected, the ranking results of the audio quality of multiple audios to be detected are determined.Since the audio frequency characteristics of audio can obtain, and audio frequency characteristics can identify the audio quality of audio to be detected, so can be according to the audio frequency characteristics of each audio to be detected, accurately determine the audio quality of multiple audios to be detected, and then accurately determine the ranking results of the audio quality of multiple audios to be detected, so as to accurately detect the highest audio of audio quality.
Description
Technical field
This application involves audio signal processing technique fields, more particularly to a kind of detection method of audio quality, device, electronics
Equipment and storage medium.
Background technique
With flourishing for Internet technology, more and more multimedia messages can spread through the internet, Yong Huke
Watch these multimedia messages with downloading or online.Wherein, audio is the multimedia messages that people often download or listen to online.
In order to provide the user with better service, the network platform for providing audio needs to be determined the quality of audio,
Tell the quality of audio quality.For example, audio distribution platform or K library, generally require the sound with more high tone quality version
Frequency updates song library, it usually needs the audio for selecting sound quality best in the audio of several versions.
Current audio quality detection method is main are as follows: by comparing detection sample corresponding with its lossless sample based on
The difference of Auditory Perception obtains a score.The higher expression sample to be tested of score and lossless Sample Similarity are higher, i.e. audio matter
It measures higher.But the lossless sample of practical application sound intermediate frequency is not all obtainable, and the difference based on Auditory Perception obtains
It is also to be grossly inaccurate to a score.
Summary of the invention
To overcome the problems in correlation technique, the embodiment of the present application provides a kind of detection method of audio quality, dress
It sets, electronic equipment and storage medium.Specific technical solution is as follows:
According to the embodiment of the present application in a first aspect, providing a kind of detection method of audio quality, which comprises
Obtain multiple audios to be detected;
Extract the audio frequency characteristics of each audio to be detected, wherein the audio frequency characteristics are to identify the acoustic to be checked
The feature of the audio quality of frequency;
According to the audio frequency characteristics of each audio to be detected, the row of the audio quality of the multiple audio to be detected is determined
Sequence result.
As an implementation, the step of audio frequency characteristics for extracting each audio to be detected, comprising:
Each audio to be detected is converted into frequency-region signal;
Calculate the effective spectrum bandwidth and/or frequency spectrum smoothness of the corresponding frequency-region signal of each audio to be detected;
The result being calculated is determined as to the audio frequency characteristics of each audio to be detected.
As an implementation, the effective spectrum band for calculating the corresponding frequency-region signal of each audio to be detected
It is wide, comprising:
For the corresponding frequency-region signal of each audio to be detected, the energy of each frequency band of the frequency-region signal is calculated
Amount;
Difference between the energy of frequency band after the energy and n-th of frequency band of n-th of frequency band persistently reaches preset value
When, determine that first frequency band to n-th of frequency band is effective band;
The effective spectrum bandwidth Y of the frequency-region signal is calculated according to formula Y=(fs/2)/K × n;
Wherein, fs is the sample rate of the audio to be detected, and K is the number of frequency bands of the frequency-region signal.
As an implementation, the frequency spectrum for calculating the corresponding frequency-region signal of each audio to be detected is steady
Degree, comprising:
For the corresponding frequency-region signal of each audio to be detected, the energy of each frequency band of the frequency-region signal is calculated
Amount;
Calculate the covariance value between the energy of every two nearby frequency bands;
The average value for calculating the covariance value, the frequency spectrum as the corresponding frequency-region signal of the audio to be detected are steady
Degree.
As an implementation, it is described calculate every two nearby frequency bands energy between covariance value the step of it
Before, the method also includes:
Time domain sliding average is carried out to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p);
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As an implementation, the multiple audio to be detected is stereo audio;
Before the step of result that will be calculated is determined as the audio frequency characteristics of each audio to be detected, institute
State method further include:
The Stereo-width for calculating the corresponding frequency-region signal of each audio to be detected, obtains calculated result.
As an implementation, the Stereo-width for calculating the corresponding frequency-region signal of each audio to be detected
The step of, comprising:
For the corresponding frequency-region signal of each audio to be detected, the left sound of each frequency band of the frequency-region signal is determined
Road signal and right-channel signals;
Calculate the association between the energy parameter of the corresponding left channel signals of each frequency band and the energy parameter of right-channel signals
Variance yields;
The average value for calculating the covariance value, the stereo width as the corresponding frequency-region signal of the audio to be detected
Degree.
As an implementation, in the stereo width for calculating the corresponding frequency-region signal of each audio to be detected
Before the step of spending, the method also includes:
Time domain sliding average is carried out to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p);
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As an implementation, the audio frequency characteristics according to each audio to be detected, determine it is the multiple to
The step of detecting the ranking results of the audio quality of audio, comprising:
The audio frequency characteristics of each audio to be detected are normalized according to preset rules respectively, obtain corresponding number
Value;
Default weight is distributed for each numerical value, and calculates the corresponding weighted value of each audio to be detected;
According to the corresponding weighted value of each audio to be detected, the audio quality of the multiple audio to be detected is determined
Ranking results.
According to the second aspect of the embodiment of the present application, a kind of detection device of audio quality is provided, described device includes:
Audio to be detected obtains module, is configured as obtaining multiple audios to be detected;
Audio feature extraction module is configured as extracting the audio frequency characteristics of each audio to be detected, wherein the sound
Frequency feature is to identify the feature of the audio quality of the audio to be detected;
Ranking results determining module is configured as being determined described more according to the audio frequency characteristics of each audio to be detected
The ranking results of the audio quality of a audio to be detected.
As an implementation, the audio feature extraction module includes:
Signal transform subblock is configured as each audio to be detected being converted to frequency-region signal;
Feature calculation submodule is configured as calculating the effective spectrum of the corresponding frequency-region signal of each audio to be detected
Bandwidth and/or frequency spectrum smoothness;
Audio frequency characteristics determine submodule, and the result for being configured as to be calculated is determined as each audio to be detected
Audio frequency characteristics.
As an implementation, the feature calculation submodule includes:
First energy calculation unit is configured as frequency-region signal corresponding for each audio to be detected, calculates institute
State the energy of each frequency band of frequency-region signal;
Effective band determination unit is configured as the energy of the frequency band after energy and n-th of frequency band of n-th of frequency band
Between difference when persistently reaching preset value, determine that first frequency band to n-th of frequency band is effective band;
Effective frequency belt width effective spectrum bandwidth calculation unit is configured as calculating frequency according to formula Y=(fs/2)/K × n
The effective spectrum bandwidth Y of domain signal;
Wherein, fs is the sample rate of the audio to be detected, and K is the number of frequency bands of the frequency-region signal.
As an implementation, the feature calculation submodule includes:
Second energy calculation unit is configured as frequency-region signal corresponding for each audio to be detected, calculates institute
State the energy of each frequency band of frequency-region signal;
Covariance value computing unit is configured as calculating the covariance value between the energy of every two nearby frequency bands;
Frequency spectrum smoothness computing unit is configured as calculating the average value of the covariance value, as the acoustic to be checked
Frequently the frequency spectrum smoothness of corresponding frequency-region signal.
As an implementation, described device further include:
First time domain sliding average module is configured as the association side between the energy for calculating every two nearby frequency bands
Before difference, time domain sliding average is carried out to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p);
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As an implementation, the multiple audio to be detected is stereo audio;
Described device further include:
Stereo-width computing module is configured as being determined as in the result that will be calculated each described to be detected
Before the audio frequency characteristics of audio, the Stereo-width of the corresponding frequency-region signal of each audio to be detected is calculated, is calculated
As a result.
As an implementation, the Stereo-width computing module includes:
Energy balane signal determines submodule, is configured as being directed to the corresponding frequency-region signal of each audio to be detected, determine
The left channel signals and right-channel signals of each frequency band of the frequency-region signal;
Covariance value computational submodule is configured as calculating the energy parameter of the corresponding left channel signals of each frequency band and the right side
Covariance value between the energy parameter of sound channel signal;
Stereo-width computational submodule is configured as calculating the average value of the covariance value, as described to be detected
The Stereo-width of the corresponding frequency-region signal of audio.
As an implementation, described device further include:
Second time domain sliding average module is configured as calculating the corresponding frequency domain letter of each audio to be detected described
Number Stereo-width before, according to formula s (t) *=s (t) × p+s (t-1) × (1-p) to the frequency-region signal carry out time domain
Sliding average;
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As an implementation, the ranking results determining module includes:
Normalize submodule, be configured as by the audio frequency characteristics of each audio to be detected respectively according to preset rules into
Row normalization, obtains corresponding numerical value;
Weighted value determines submodule, is configured as distributing default weight for each numerical value, and calculate each described to be detected
The corresponding weighted value of audio;
Ranking results determine submodule, are configured as determining institute according to the corresponding weighted value of each audio to be detected
State the ranking results of the audio quality of multiple audios to be detected.
According to the third aspect of the embodiment of the present application, a kind of electronic equipment, including processor, communication interface, storage are provided
Device and communication bus, wherein the processor, the communication interface, the memory are completed each other by communication bus
Communication;
The memory, for storing computer program;
The processor when for executing the program stored on the memory, realizes any of the above-described audio matter
The detection method step of amount.
According to the fourth aspect of the embodiment of the present application, a kind of non-transitorycomputer readable storage medium is provided, when described
When instruction in storage medium is executed by the processor of electronic equipment so that the electronic equipment be able to carry out it is any of the above-described described
The detection method step of audio quality.
In scheme provided by the embodiment of the present application, then the available multiple audios to be detected of electronic equipment are extracted every
The audio frequency characteristics of a audio to be detected, audio frequency characteristics are the feature for identifying the audio quality of audio to be detected, in turn, according to each
The audio frequency characteristics of audio to be detected determine the ranking results of the audio quality of multiple audios to be detected.Since the audio of audio is special
Sign can obtain, and audio frequency characteristics can identify the audio quality of audio to be detected, it is possible to according to each to be checked
The audio frequency characteristics of acoustic frequency, accurately determine the audio quality of multiple audios to be detected, and then accurately determine multiple acoustics to be checked
The ranking results of the audio quality of frequency, so as to accurately detect the highest audio of audio quality.More than it should be understood that
General description and following detailed description be only it is exemplary and explanatory, the application can not be limited.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the application
Example, and together with specification it is used to explain the principle of the application.
Fig. 1 is a kind of flow chart of the detection method of audio quality shown according to an exemplary embodiment;
Fig. 2 is a kind of specific flow chart of step S102 in embodiment illustrated in fig. 1;
Fig. 3 is a kind of flow chart of calculation of effective spectrum bandwidth shown according to an exemplary embodiment;
Fig. 4 is a kind of flow chart of calculation of frequency spectrum smoothness shown according to an exemplary embodiment;
Fig. 5 is a kind of flow chart of calculation of Stereo-width shown according to an exemplary embodiment
Fig. 6 is a kind of specific flow chart of step S103 in embodiment illustrated in fig. 1;
Fig. 7 is a kind of detection device block diagram of audio quality shown according to an exemplary embodiment;
Fig. 8 is a kind of electronic equipment block diagram shown according to an exemplary embodiment;
Fig. 9 is a kind of specific block diagram of electronic equipment in embodiment illustrated in fig. 8.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the application.
In order to realize the detection of audio quality, and the accuracy of audio quality detection is improved, the embodiment of the present application provides
A kind of detection method of audio quality, device, electronic equipment and non-transitorycomputer readable storage medium.
A kind of detection method of audio quality provided by the embodiment of the present application is introduced first below.
A kind of detection method of audio quality provided by the embodiment of the present application can be applied to any need to detect audio
Electronic equipment of quality, such as processor, mobile phone, computer etc., are not specifically limited herein.For the convenience of description, hereinafter referred to as
Electronic equipment.
As shown in Figure 1, a kind of detection method of audio quality, the method includes the steps S101- step S103:
In step s101, multiple audios to be detected are obtained;
In step s 102, the audio frequency characteristics of each audio to be detected are extracted, wherein the audio frequency characteristics are mark
The feature of the audio quality of the audio to be detected;
In step s 103, according to the audio frequency characteristics of each audio to be detected, the multiple audio to be detected is determined
Audio quality ranking results.
As it can be seen that in scheme provided by the embodiment of the present application, then the available multiple audios to be detected of electronic equipment mention
The audio frequency characteristics of each audio to be detected are taken, audio frequency characteristics are the feature for identifying the audio quality of audio to be detected, in turn, according to
The audio frequency characteristics of each audio to be detected determine the ranking results of the audio quality of multiple audios to be detected.Due to the sound of audio
Frequency feature can obtain, and audio frequency characteristics can identify the audio quality of audio to be detected, it is possible to according to each
The audio frequency characteristics of audio to be detected, accurately determine the audio quality of multiple audios to be detected, and then accurately determine multiple to be checked
The ranking results of the audio quality of acoustic frequency, so as to accurately detect the highest audio of audio quality.
In above-mentioned steps S101, the available multiple audios to be detected of electronic equipment, multiple audio to be detected is
Need to detect the audio of audio quality.Audio to be detected can be the audio that electronic equipment is locally stored, or from network
Or the audio that other equipment obtain, this is all reasonably, to be not specifically limited herein.
For example, need to determine the best audio of audio quality from the audio of several versions when needing to update song library, into
And song library is updated using the best audio of the audio quality, then the audio of several versions is above-mentioned audio to be detected.
After obtaining above-mentioned audio to be detected, in order to determine that the audio quality of each audio to be detected, electronic equipment can extract
The audio frequency characteristics of each audio to be detected.Wherein, audio frequency characteristics are the feature for identifying the audio quality of audio to be detected, for example,
It can be effective spectrum bandwidth, frequency spectrum smoothness, Stereo-width etc., be not specifically limited herein.
Since audio frequency characteristics can identify the audio quality of each audio to be detected, thus electronic equipment can execute it is above-mentioned
Step S103 determines the sequence knot of the audio quality of multiple audios to be detected that is, according to the audio frequency characteristics of each audio to be detected
Fruit.
In many application scenarios, the absolute quality for determining the audio quality of each audio to be detected is not needed, but
It needs to be determined that the audio quality of several audios to be detected is relatively fine or not, so that it is determined that the wherein optimal acoustic to be checked of audio quality
Frequently, multiple acoustics to be checked can rapidly and accurately be determined using the detection method of audio quality provided by the embodiment of the present application
The ranking results of the audio quality of frequency, also can the wherein optimal audio to be detected of audio quality.
As a kind of embodiment of the embodiment of the present application, as shown in Fig. 2, the audio of each audio to be detected of said extracted
The step of feature may include:
Each audio to be detected is converted to frequency-region signal by step S201;
In order to facilitate the audio frequency characteristics of each audio to be detected of determination, each audio to be detected can be converted to frequency first
Domain signal.Each audio to be detected can be converted into frequency-region signal using modes such as Fourier transformation, Fast Fourier Transform (FFT)s.
Step S202 calculates the effective spectrum bandwidth and/or frequency spectrum of the corresponding frequency-region signal of each audio to be detected
Smoothness;
The result being calculated is determined as the audio frequency characteristics of each audio to be detected by step S203.
In one embodiment, after each audio to be detected is converted to frequency-region signal by electronic equipment, frequency-region signal tool
There are multiple frequency bands, and due to the influence of the factors such as noise, the effective spectrum bandwidth of audio usually all can be lower than theoretical effective spectrum
Bandwidth, part low-down for frequency band energy may be considered invalid frequency band, therefore effective spectrum bandwidth can identify audio
The quality of quality, so electronic equipment can calculate the effective spectrum bandwidth of the corresponding frequency-region signal of each audio to be detected, it will
The effective spectrum bandwidth of the corresponding frequency-region signal of each audio to be detected is determined as the audio frequency characteristics of each audio to be detected.
In another embodiment, theoretically the corresponding frequency-region signal of audio should be continuously without recess on frequency spectrum
, and due to causing the corresponding frequency-region signal of audio to have the place being much recessed on frequency spectrum by factors such as encoder compresses,
Compression ratio is bigger, and recess also may be more, and the audio quality of audio is also poorer, so frequency spectrum smoothness can identify audio matter
The quality of amount, so electronic equipment can calculate the frequency spectrum smoothness of the corresponding frequency-region signal of each audio to be detected, it will be each
The frequency spectrum smoothness of the corresponding frequency-region signal of audio to be detected is determined as the audio frequency characteristics of each audio to be detected.
Certainly, electronic equipment can also be by the effective spectrum bandwidth and frequency spectrum of the corresponding frequency-region signal of each audio to be detected
Smoothness is determined as the audio frequency characteristics of each audio to be detected, this is all reasonably, to be not specifically limited herein.
As it can be seen that in the present embodiment, each audio to be detected can be converted to frequency-region signal by electronic equipment, then calculate
The effective spectrum bandwidth and/or frequency spectrum smoothness of the corresponding frequency-region signal of each audio to be detected, and then the knot that will be calculated
Fruit is determined as the audio frequency characteristics of each audio to be detected.It can rapidly and accurately determine the audio frequency characteristics of each audio to be detected,
Convenient for the ranking results of the audio quality of the multiple audios to be detected of subsequent determination.
As a kind of embodiment of the embodiment of the present application, as shown in figure 3, each audio pair to be detected of above-mentioned calculating
The step of effective spectrum bandwidth for the frequency-region signal answered, may include:
Step S301 calculates each of described frequency-region signal for the corresponding frequency-region signal of each audio to be detected
The energy of frequency band;
The corresponding frequency-region signal of each audio to be detected all has multiple frequency bands, in order to determine that each audio to be detected is corresponding
Frequency-region signal effective spectrum bandwidth, electronic equipment can calculate the energy of each frequency band of frequency-region signal.Wherein, frequency is calculated
The concrete mode of the energy of each frequency band of domain signal can be using the frequency band of any calculating frequency-region signal of field of audio processing
Energy mode, be not specifically limited and illustrate herein.
Step S302, the difference between the energy of the frequency band after energy and n-th of frequency band of n-th of frequency band last up to
When to preset value, determine that first frequency band to n-th of frequency band is effective band;
In frequency-region signal, as the energy of the raising frequency band of frequency band gradually decreases, and energy it is very low frequency band it is corresponding
Signal is invalid.So if the difference of the energy of the energy and first frequency band after n-th of frequency band of n-th frequency band reaches
To preset value, illustrate that the corresponding signal of (n+1)th frequency band may be invalid.
If the difference between the energy of n-th frequency band and the energy of the frequency band after n-th of frequency band persistently reaches pre-
If value, so that it may determine the frequency band after n-th of frequency band without effective audio power, corresponding signal i.e. nothing
Imitate signal.So electronic equipment can determine that first frequency band to n-th of frequency band is effective band.
Wherein, above-mentioned preset value can be determined according to factors such as the energy of actual frequency domain signal, for example, can be 60 points
Shellfish, 70 decibels, 75 decibels etc., are not specifically limited herein.
Step S303 calculates the effective spectrum bandwidth Y of the frequency-region signal according to formula Y=(fs/2)/K × n.
After above-mentioned effective band has been determined, electronic equipment can calculate frequency domain letter according to formula Y=(fs/2)/K × n
Number effective spectrum bandwidth Y.Wherein, fs is the sample rate of audio to be detected, and K is the number of frequency bands of frequency-region signal.Fs/2 is indicated
The theoretical effective spectrum bandwidth of audio to be detected, so when the effective band of frequency-region signal is n, effective frequency of frequency-region signal
The wide Y of bands of a spectrum is (fs/2)/K × n.
As it can be seen that in the present embodiment, electronic equipment can be directed to the corresponding frequency-region signal of each audio to be detected, frequency is calculated
The energy of each frequency band of domain signal, the difference between the energy of the frequency band after energy and n-th of frequency band of n-th of frequency band
When persistently reaching preset value, determine that first frequency band to n-th of frequency band is effective band, and then according to formula Y=(fs/
2)/K × n calculates the effective spectrum bandwidth Y of frequency-region signal.In this way, the effective spectrum bandwidth of frequency-region signal can accurately be determined, really
Definitely determine the audio frequency characteristics of each audio to be detected.
As a kind of embodiment of the embodiment of the present application, as shown in figure 4, each audio pair to be detected of above-mentioned calculating
The step of frequency spectrum smoothness for the frequency-region signal answered, may include:
Step S401 calculates each of described frequency-region signal for the corresponding frequency-region signal of each audio to be detected
The energy of frequency band;
The mode for calculating the energy of each frequency band of frequency-region signal is identical as above-mentioned steps S301, and details are not described herein.
Step S402 calculates the covariance value between the energy of every two nearby frequency bands;
After obtaining the energy of each frequency band of frequency-region signal, electronic equipment can calculate the energy of every two nearby frequency bands
Between covariance value.For convenience, it is illustrated so that frequency-region signal has 5 frequency bands as an example.For example, 5 frequency bands
For frequency band 1, frequency band 2, frequency band 3, frequency band 4 and frequency band 5, then electronic equipment can calculate between frequency band 1 and the energy of frequency band 2
Covariance value a, the covariance value b between frequency band 2 and the energy of frequency band 3, the covariance between frequency band 3 and the energy of frequency band 4
Covariance value d between value c, frequency band 4 and the energy of frequency band 5.
If audio to be detected is stereo audio, the corresponding frequency-region signal of audio to be detected is respectively L channel pair
The corresponding frequency-region signal of frequency-region signal and right channel answered, then electronic equipment calculates between the energy of every two nearby frequency bands
When covariance value, the covariance between the energy of the corresponding frequency-region signal every two nearby frequency bands of L channel can be calculated separately
Covariance value between value and the energy of the corresponding frequency-region signal every two nearby frequency bands of right channel.
What covariance indicated is the overall error of two variables, if the variation tendency of two variables is consistent, that is,
It says if another is also greater than itself desired value when one of them is greater than itself desired value, between two variables
Covariance is exactly positive value;If the variation tendency of two variables is on the contrary, another when i.e. one of variable is greater than the desired value of itself
An outer desired value less than itself, then the covariance between two variables is exactly negative value, so two nearby frequency bands
Covariance value between energy is smaller, and the variation tendency difference of the energy of two nearby frequency bands is bigger, the energy of two nearby frequency bands
Correlation between amount is lower, illustrates a possibility that continuity between each frequency band of frequency-region signal is poorer, and audio quality is poor
It is bigger.
Step S403 calculates the average value of the covariance value, as the corresponding frequency-region signal of the audio to be detected
Frequency spectrum smoothness.
After above-mentioned covariance value is calculated, electronic equipment can calculate the average value of covariance value, by the average value
Frequency spectrum smoothness as the corresponding frequency-region signal of audio to be detected.For example, association is calculated in electronic equipment in step S402
Variance yields a, covariance value b, covariance value c and covariance value d.It so can be by covariance value a, covariance value b, covariance
Frequency spectrum smoothness of the average value of value c and covariance value d as frequency-region signal.In this way, can be calculated each to be detected
The corresponding frequency spectrum smoothness of audio.
As it can be seen that in the present embodiment, electronic equipment can be directed to the corresponding frequency-region signal of each audio to be detected, frequency is calculated
Then the energy of each frequency band of domain signal calculates the covariance value between the energy of every two nearby frequency bands, calculate covariance
The average value of value, using the average value as the frequency spectrum smoothness of the corresponding frequency-region signal of audio to be detected.In this way, can accurately really
Determine the frequency spectrum smoothness of frequency-region signal, it is ensured that accurately determine the audio frequency characteristics of each audio to be detected.
Random perturbation due to audio etc. may influence the judgement of audio quality, so in order to mitigate the random perturbation of audio
Influence to the frequency spectrum smoothness for calculating the corresponding frequency-region signal of audio to be detected, a kind of embodiment party as the embodiment of the present application
Before the step of formula, covariance value between the energy of above-mentioned calculating every two nearby frequency bands, the above method can also include:
Time domain sliding average is carried out to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p).
Wherein, frequency-region signal corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
Energy, s (t) be audio to be detected in t moment audio frame corresponding to frequency-region signal energy, s (t-1) be acoustic to be checked
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in frequency, p are default slide coefficient, p ∈ (0,1).
Audio to be detected generally comprises multiframe audio frame, and every frame audio frame can obtain corresponding frequency-region signal by transformation,
Electronic equipment can carry out time domain sliding average to these frequency-region signals, to mitigate the influence of the random perturbation of audio.
Specifically, electronic equipment can carry out time domain sliding average to frequency-region signal using above-mentioned formula.Wherein, p is
The specific value of the default slide coefficient being worth between 0 to 1, p can be determining according to factors such as the energy of frequency-region signal, herein not
It is specifically limited, for example, can be 0.2,0.5,0.8 etc..
As it can be seen that in the present embodiment, the step of covariance value between the energy for calculating every two nearby frequency bands before,
It is flat that electronic equipment can carry out time domain sliding to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p)
, to mitigate the influence of the random perturbation of audio, keep the frequency spectrum smoothness for the frequency-region signal being calculated more accurate.
As a kind of embodiment of the embodiment of the present application, above-mentioned multiple audios to be detected can be stereo audio.Needle
For the case where audio to be detected is stereo audio, it is determined as each audio to be detected in the above-mentioned result that will be calculated
Audio frequency characteristics the step of before, the above method can also include:
The Stereo-width for calculating the corresponding frequency-region signal of each audio to be detected, obtains calculated result.
Since for stereo audio, Stereo-width is an important index, audio quality can be characterized
Quality, so electronic equipment can calculate the Stereo-width of the corresponding frequency-region signal of each audio to be detected, this is stereo
Width is also used as the audio frequency characteristics of audio to be detected.Stereo-width is wider, shows that the stereophonic effect of audio to be detected is better,
Audio quality is better;Stereo-width is narrower, shows that the stereophonic effect of audio to be detected is poorer, audio quality is poorer.
In one embodiment, before the Stereo-width for calculating the corresponding frequency-region signal of each audio to be detected,
Time domain sliding average can also be carried out to above-mentioned frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p), to subtract
The influence of the random perturbation of schwa frequency guarantees the accuracy of the Stereo-width of frequency-region signal.Specific time domain sliding average mode
It is introduced in the above-described embodiments, details are not described herein.
As it can be seen that in the present embodiment, when audio to be detected is stereo audio, being determined as by the result being calculated
Before the step of audio frequency characteristics of each audio to be detected, electronic equipment can calculate the corresponding frequency domain letter of each audio to be detected
Number Stereo-width, obtain calculated result.And then Stereo-width is also used as to the audio frequency characteristics of audio to be detected, so as to
The audio frequency characteristics for detecting audio are more comprehensive, can accurately characterize the audio quality of audio to be detected.
As a kind of embodiment of the embodiment of the present application, as shown in figure 5, each audio to be detected of above-mentioned calculating is corresponding
The step of Stereo-width of frequency-region signal, may include:
Step S501 determines each of described frequency-region signal for the corresponding frequency-region signal of each audio to be detected
The left channel signals and right-channel signals of frequency band;
For stereo audio, the frequency-region signal of L channel and the frequency of right channel can be obtained by being transformed to frequency-region signal
Domain signal.For the corresponding frequency-region signal of each audio to be detected, electronic equipment can calculate separately the every of frequency-region signal first
The L channel frequency spectrum and right channel frequency spectrum of a frequency band also just obtain the left channel signals and right channel of each frequency band of frequency-region signal
Signal.Wherein, the mode for specifically calculating L channel frequency spectrum and right channel frequency spectrum can be using in Fourier transformation or quick Fu
Leaf transformation or other time-domain signals are converted to the mode of frequency-region signal.
Step S502 calculates the energy parameter of the corresponding left channel signals of each frequency band and the energy parameter of right-channel signals
Between covariance value;
After obtaining the corresponding left channel signals of each frequency band and right-channel signals, electronic equipment can calculate each frequency band
Covariance value between the energy parameter of corresponding left channel signals and the energy parameter of right-channel signals.Wherein, energy parameter
Can be energy or phase and other can indicate the parameter of signal energy, this is all reasonable.
It is illustrated so that the corresponding L channel of frequency-region signal has 5 frequency bands as an example.For example, audio S to be detected is corresponding
There is L channel frequency-region signal a frequency band left side 1, a frequency band left side 2, a frequency band left side 3, a frequency band left side 4 and a frequency band left side 5, right channel frequency-region signal to have
The frequency band right side 1, the frequency band right side 2, the frequency band right side 3, the frequency band right side 4 and the frequency band right side 5, then electronic equipment can calculate a frequency band left side 1 and the frequency band right side 1
Signal energy parameter between covariance value A, frequency band a left side 2 and the frequency band right side 2 signal energy parameter between covariance
Covariance value C between the energy parameter of the signal on value B, a frequency band left side 3 and the frequency band right side 3, a frequency band left side 4 and the signal on the frequency band right side 4
Covariance value E between the energy parameter of the signal of covariance value D and a frequency band left side 5 and the frequency band right side 5 between energy parameter.
What covariance indicated is the overall error of two variables, if the variation tendency of two variables is consistent, that is,
It says if another is also greater than itself desired value when one of them is greater than itself desired value, between two variables
Covariance is exactly positive value;If the variation tendency of two variables is on the contrary, another when i.e. one of variable is greater than the desired value of itself
An outer desired value less than itself, then the covariance between two variables is exactly negative value, so the energy of left channel signals
The covariance value measured between parameter and the energy parameter of right-channel signals is smaller, and the variation of left channel signals and right-channel signals becomes
For potential difference away from bigger, the correlation between left channel signals and right-channel signals is lower, illustrates the Stereo-width of audio to be detected
It is wider;Covariance value between the energy parameter of left channel signals and the energy parameter of right-channel signals is bigger, left channel signals
Closer with the variation tendency of right-channel signals, the correlation between left channel signals and right-channel signals is higher, illustrates to be checked
A possibility that Stereo-width of acoustic frequency is narrower, closer monophonic signal, audio quality is poor is also bigger.
Step S503 calculates the average value of the covariance value, as the corresponding frequency-region signal of the audio to be detected
Stereo-width.
After the covariance value between the corresponding left channel signals of each frequency band and right-channel signals is calculated, it can count
The average value for calculating all covariance values, using the average value as the Stereo-width of the corresponding frequency-region signal of audio to be detected.
For example, covariance value A- covariance value E is calculated in electronic equipment, then electronic equipment is just in step S502
The average value that covariance value A- covariance value E can be calculated, using the average value as the corresponding frequency-region signal of audio S to be detected
Stereo-width.
As it can be seen that in the present embodiment, electronic equipment can be directed to the corresponding frequency-region signal of each audio to be detected, frequency is calculated
The left channel signals and right-channel signals of each frequency band of domain signal, then calculate the energy of the corresponding left channel signals of each frequency band
The covariance value between parameter and the energy parameter of right-channel signals is measured, and then calculates the average value of covariance value, as to be checked
The Stereo-width of the corresponding frequency-region signal of acoustic frequency.In this way, the Stereo-width of frequency-region signal can be determined accurately, it is ensured that quasi-
Determine the audio frequency characteristics of each audio to be detected.
As a kind of embodiment of the embodiment of the present application, as shown in fig. 6, above-mentioned according to each audio to be detected
Audio frequency characteristics the step of determining the ranking results of the audio quality of the multiple audio to be detected, may include:
The audio frequency characteristics of each audio to be detected are normalized according to preset rules respectively, obtain by step S601
To corresponding numerical value;
In order to facilitate the ranking results of the audio quality of determination audio to be detected, electronic equipment can be according to preset rules pair
The numerical value of the audio frequency characteristics of each audio to be detected is normalized.In a kind of mode, the numberical range that normalizes is-
1 to 1, closer -1, indicate that audio quality is better;Closer to 1, indicate that audio quality is poorer.
For example, the audio frequency characteristics of each audio to be detected include above-mentioned effective spectrum bandwidth, frequency spectrum smoothness and stereo
Width, then, since the numerical value of effective spectrum bandwidth, frequency spectrum smoothness and Stereo-width is bigger, indicate that audio quality is better,
So electronic equipment can be normalized effective spectrum bandwidth, frequency spectrum smoothness and Stereo-width respectively, effective spectrum
The numerical value of bandwidth, frequency spectrum smoothness and Stereo-width is bigger, and corresponding normalization result is smaller.
Step S602 distributes default weight for each numerical value, and calculates the corresponding weighted value of each audio to be detected;
Since each feature that audio frequency characteristics include has difference in the importance of characterization audio quality, so electronics is set
Standby can be that each numerical value distributes default weight appropriate according to the importance of each characteristic present audio quality.Characterize audio matter
The importance of amount is higher, then can distribute higher weight;The importance for characterizing audio quality is lower, then can distribute lower
Weight.
For example, it is assumed that the audio frequency characteristics of each audio to be detected include above-mentioned effective spectrum bandwidth, frequency spectrum smoothness and stand
The importance of body sound width, effective spectrum bandwidth, frequency spectrum smoothness and Stereo-width characterization audio quality successively reduces, then
Weight 0.5,0.3 and 0.2 can be distributed respectively for effective spectrum bandwidth, frequency spectrum smoothness and Stereo-width.
In one embodiment, since in some scenes, effective spectrum bandwidth and Stereo-width are for audio matter
The influence of amount is very big, therefore, effective spectrum bandwidth threshold and Stereo-width threshold value can be set, if effective spectral bandwidth
Not up to effective spectrum bandwidth threshold illustrates the audio to be detected alternatively, Stereo-width is not up to Stereo-width threshold value
Audio quality is poor, can set the corresponding default weight of the effective spectrum bandwidth and Stereo-width of the audio to be detected to
0。
After each numerical value that audio frequency characteristics for audio to be detected include distributes default weight, it can calculate each to be checked
The corresponding weighted value of acoustic frequency.For example, the numerical value that the audio frequency characteristics of audio A to be detected include is -0.8, -0.5 and 0.1, it is corresponding
Default weight be respectively 0.8,0.1 and 0.1, then corresponding -0.8 × 0.8+ of weighted value (- 0.5) × 0.1 of audio A to be detected
+ 0.1 × 0.1=-0.68.
Step S603 determines the multiple audio to be detected according to the corresponding weighted value of each audio to be detected
The ranking results of audio quality.
After obtaining the corresponding weighted value of each audio to be detected, the audio quality of multiple audios to be detected can be determined
Ranking results.If when audio frequency characteristics are normalized, according to preset rules be numberical range be -1 to 1,
Closer -1, it indicates that audio quality is better, then weighted value is smaller, indicates that the audio quality of corresponding audio to be detected is higher.
If when audio frequency characteristics are normalized, according to the preset rules numberical range that is be -1 to 1, closer to 1, table
Show that audio quality is better, then weighted value is bigger, indicates that the audio quality of corresponding audio to be detected is higher.
In one embodiment, if audio to be detected is two, which acoustic to be checked is distinguished for convenience
The audio quality of frequency is more preferable, and certain rule can also be arranged in electronic equipment, and the corresponding weighted value of two audios to be detected is carried out
It calculates, obtains a numerical value, determine that the audio quality of which audio to be detected in audio to be detected is more preferable according to the numerical value.
For example, if when audio frequency characteristics are normalized, according to the preset rules numberical range that is be -1
To 1, closer -1, indicate that audio quality is better, then electronic equipment can by the corresponding weighted value of audio m to be detected with it is to be checked
The corresponding weighted value of acoustic frequency n subtracts each other to obtain a numerical value t, and is three subintervals by -1 to 1 interval division, respectively [- 1,
A], [- a, a] and [a, 1], if t belongs to subinterval [- 1, a], it is determined that the audio quality of audio m to be detected is more preferable;If t
Belong to subinterval [- a, a], it is determined that the audio quality of audio m to be detected and audio n to be detected are not much different;If t belongs to son
Section [a, 1], it is determined that the audio quality of audio n to be detected is more preferable.
Wherein, the occurrence of a can be determined according to the comparison result of a large amount of audios, in order to keep the value of a more accurate, sound
The ranking results of frequency quality are more accurate, and electronic equipment can use the sample manually marked and be verified, and adjust taking for a
Value, so that the ranking results of the audio quality of audio to be detected are more accurate.
As it can be seen that in the present embodiment, electronic equipment can be by the audio frequency characteristics of each audio to be detected according to preset rules
Be normalized, obtain corresponding numerical value, distribute default weight for each numerical value, and calculate each audio to be detected it is corresponding plus
Weight according to the corresponding weighted value of each audio to be detected, determines the row of the audio quality of the multiple audio to be detected in turn
Sequence result.In this way, can rapidly and accurately determine the ranking results of the audio quality of audio to be detected, the standard of audio quality detection
Exactness is high.
Fig. 7 is a kind of detection device block diagram of audio quality shown according to an exemplary embodiment.As shown in fig. 7, one
The detection device of kind audio quality, described device include:
Audio to be detected obtains module 710, is configured as obtaining multiple audios to be detected;
Audio feature extraction module 720 is configured as extracting the audio frequency characteristics of each audio to be detected;
Wherein, the audio frequency characteristics are the feature for identifying the audio quality of the audio to be detected.
Ranking results determining module 730, is configured as the audio frequency characteristics according to each audio to be detected, determine described in
The ranking results of the audio quality of multiple audios to be detected.
As it can be seen that in scheme provided by the embodiment of the present application, then the available multiple audios to be detected of electronic equipment mention
The audio frequency characteristics of each audio to be detected are taken, audio frequency characteristics are the feature for identifying the audio quality of audio to be detected, in turn, according to
The audio frequency characteristics of each audio to be detected determine the ranking results of the audio quality of multiple audios to be detected.Due to the sound of audio
Frequency feature can obtain, and audio frequency characteristics can identify the audio quality of audio to be detected, it is possible to according to each
The audio frequency characteristics of audio to be detected, accurately determine the audio quality of multiple audios to be detected, and then accurately determine multiple to be checked
The ranking results of the audio quality of acoustic frequency, so as to accurately detect the highest audio of audio quality.
As a kind of embodiment of the embodiment of the present application, above-mentioned audio feature extraction module 720 may include:
Signal transform subblock (is not shown) in Fig. 7, is configured as being converted to each audio to be detected into frequency domain letter
Number;
Feature calculation submodule (is not shown) in Fig. 7, is configured as calculating the corresponding frequency domain of each audio to be detected
The effective spectrum bandwidth and/or frequency spectrum smoothness of signal;
Audio frequency characteristics determine submodule (being not shown in Fig. 7), the result for being configured as to be calculated be determined as it is each to
Detect the audio frequency characteristics of audio.
As a kind of embodiment of the embodiment of the present application, features described above computational submodule may include:
First energy calculation unit (is not shown) in Fig. 7, is configured as frequency corresponding for each audio to be detected
Domain signal calculates the energy of each frequency band of the frequency-region signal;
Effective band determination unit (is not shown) in Fig. 7, be configured as when the energy of n-th frequency band and n-th frequency band it
When difference between the energy of frequency band afterwards persistently reaches preset value, determine that first frequency band to n-th of frequency band is effective
Frequency band;
Effective spectrum bandwidth calculation unit (is not shown) in Fig. 7, is configured as being calculated according to formula Y=(fs/2)/K × n
The effective spectrum bandwidth Y of frequency-region signal;
Wherein, fs is the sample rate of the audio to be detected, and K is the number of frequency bands of the frequency-region signal.
As a kind of embodiment of the embodiment of the present application, features described above computational submodule may include:
Second energy calculation unit (is not shown) in Fig. 7, is configured as frequency corresponding for each audio to be detected
Domain signal calculates the energy of each frequency band of the frequency-region signal;
Covariance value computing unit (is not shown) in Fig. 7, is configured as between the energy for calculating every two nearby frequency bands
Covariance value;
Frequency spectrum smoothness computing unit (being not shown in Fig. 7) is configured as calculating the average value of the covariance value, as
The frequency spectrum smoothness of the corresponding frequency-region signal of the audio to be detected.
As a kind of embodiment of the embodiment of the present application, above-mentioned apparatus can also include:
First time domain sliding average module (being not shown in Fig. 7) is configured as in the calculating every two nearby frequency bands
Before covariance value between energy, the frequency-region signal is carried out according to formula s (t) *=s (t) × p+s (t-1) × (1-p)
Time domain sliding average;
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As a kind of embodiment of the embodiment of the present application, above-mentioned multiple audios to be detected can be stereo audio;
Above-mentioned apparatus can also include:
Stereo-width computing module (is not shown) in Fig. 7, is configured as being determined as in the result that will be calculated
Before the audio frequency characteristics of each audio to be detected, the stereo of the corresponding frequency-region signal of each audio to be detected is calculated
Width obtains calculated result.
As a kind of embodiment of the embodiment of the present application, above-mentioned Stereo-width computing module may include:
Signal determines submodule (being not shown in Fig. 7), is configured as being directed to the corresponding frequency-region signal of each audio to be detected,
Determine the left channel signals and right-channel signals of each frequency band of the frequency-region signal;
Covariance value computational submodule (is not shown) in Fig. 7, is configured as calculating the corresponding left channel signals of each frequency band
Energy parameter and right-channel signals energy parameter between covariance value;
Stereo-width computational submodule (is not shown) in Fig. 7, is configured as calculating the average value of the covariance value, makees
For the Stereo-width of the corresponding frequency-region signal of the audio to be detected.
As a kind of embodiment of the embodiment of the present application, above-mentioned apparatus further include:
Second time domain sliding average module is configured as calculating the corresponding frequency domain letter of each audio to be detected described
Number Stereo-width before, according to formula s (t) *=s (t) × p+s (t-1) × (1-p) to the frequency-region signal carry out time domain
Sliding average;
Wherein, frequency domain corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
The energy of signal, s (t) are the energy of frequency-region signal corresponding to the audio frame of t moment in the audio to be detected, and s (t-1) is
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in the audio to be detected, p are to preset slide coefficient, p ∈ (0,
1)。
As a kind of embodiment of the embodiment of the present application, above-mentioned ranking results determining module 730 may include:
It normalizes submodule (being not shown in Fig. 7), is configured as distinguishing the audio frequency characteristics of each audio to be detected
It is normalized according to preset rules, obtains corresponding numerical value;
Weighted value determines submodule (being not shown in Fig. 7), is configured as distributing default weight for each numerical value, and calculate every
The corresponding weighted value of a audio to be detected;
Ranking results determine submodule (being not shown in Fig. 7), are configured as corresponding according to each audio to be detected
Weighted value determines the ranking results of the audio quality of the multiple audio to be detected.
The embodiment of the present application also provides a kind of electronic equipment, as shown in figure 8, electronic equipment may include processor 801,
Communication interface 802, memory 803 and communication bus 804, wherein processor 801, communication interface 802, memory 803 pass through logical
Letter bus 804 completes mutual communication,
Memory 803, for storing computer program;
Processor 801 when for executing the program stored on memory 803, is realized described in any of the above-described embodiment
The detection method step of audio quality.
As it can be seen that in scheme provided by the embodiment of the present application, then the available multiple audios to be detected of electronic equipment mention
The audio frequency characteristics of each audio to be detected are taken, audio frequency characteristics are the feature for identifying the audio quality of audio to be detected, in turn, according to
The audio frequency characteristics of each audio to be detected determine the ranking results of the audio quality of multiple audios to be detected.Due to the sound of audio
Frequency feature can obtain, and audio frequency characteristics can identify the audio quality of audio to be detected, it is possible to according to each
The audio frequency characteristics of audio to be detected, accurately determine the audio quality of multiple audios to be detected, and then accurately determine multiple to be checked
The ranking results of the audio quality of acoustic frequency, so as to accurately detect the highest audio of audio quality.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.;It can also be digital signal processor (Digital Signal
Processing, DSP), it is specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.
Fig. 9 is the block diagram of a kind of electronic equipment 900 shown according to an exemplary embodiment.For example, electronic equipment 900 can
To be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, Medical Devices are good for
Body equipment, personal digital assistant etc..
Referring to Fig. 9, electronic equipment 900 may include following one or more components: processing component 902, memory 904,
Electric power assembly 906, multimedia component 908, audio component 910, the interface 912 of input/output (I/O), sensor module 914,
And communication component 916.
The integrated operation of the usual controlling electronic devices 900 of processing component 902, such as with display, call, data are logical
Letter, camera operation and record operate associated operation.Processing component 902 may include one or more processors 920 to hold
Row instruction, to perform all or part of the steps of the methods described above.In addition, processing component 902 may include one or more moulds
Block, convenient for the interaction between processing component 902 and other assemblies.For example, processing component 902 may include multi-media module, with
Facilitate the interaction between multimedia component 908 and processing component 902.
Memory 904 is configured as storing various types of data to support the operation in electronic equipment 900.These data
Example include any application or method for being operated on electronic equipment 900 instruction, contact data, telephone directory
Data, message, picture, video etc..Memory 904 can by any kind of volatibility or non-volatile memory device or it
Combination realize, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable
Except programmable read only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, fastly
Flash memory, disk or CD.
Power supply module 906 provides electric power for the various assemblies of electronic equipment 900.Power supply module 906 may include power supply pipe
Reason system, one or more power supplys and other with for electronic equipment 900 generate, manage, and distribute the associated component of electric power.
Multimedia component 908 includes the screen of one output interface of offer between electronic equipment 900 and user.One
In a little embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touch sensings
Device is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding action
Boundary, but also detect duration and pressure associated with the touch or slide operation.In some embodiments, more matchmakers
Body component 908 includes a front camera and/or rear camera.When electronic equipment 900 is in operation mode, as shot mould
When formula or video mode, front camera and/or rear camera can receive external multi-medium data.Each preposition camera shooting
Head and rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 910 is configured as output and/or input audio signal.For example, audio component 910 includes a Mike
Wind (MIC), when electronic equipment 900 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone
It is configured as receiving external audio signal.The received audio signal can be further stored in memory 904 or via logical
Believe that component 916 is sent.In some embodiments, audio component 910 further includes a loudspeaker, is used for output audio signal.
I/O interface 912 provides interface between processing component 902 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lock
Determine button.
Sensor module 914 includes one or more sensors, for providing the state of various aspects for electronic equipment 900
Assessment.For example, sensor module 914 can detecte the state that opens/closes of electronic equipment 900, the relative positioning of component, example
As the component be electronic equipment 900 display and keypad, sensor module 914 can also detect electronic equipment 900 or
The position change of 900 1 components of electronic equipment, the existence or non-existence that user contacts with electronic equipment 900, electronic equipment 900
The temperature change of orientation or acceleration/deceleration and electronic equipment 900.Sensor module 914 may include proximity sensor, be configured
For detecting the presence of nearby objects without any physical contact.Sensor module 914 can also include optical sensor,
Such as CMOS or ccd image sensor, for being used in imaging applications.In some embodiments, which may be used also
To include acceleration transducer, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 916 is configured to facilitate the communication of wired or wireless way between electronic equipment 900 and other equipment.
Electronic equipment 900 can access the wireless network based on communication standard, such as WiFi, carrier network (such as 2G, 3G, 4G or 5G),
Or their combination.In one exemplary embodiment, communication component 916 receives via broadcast channel and comes from external broadcasting management
The broadcast singal or broadcast related information of system.In one exemplary embodiment, the communication component 916 further includes that near field is logical
(NFC) module is believed, to promote short range communication.For example, radio frequency identification (RFID) technology, infrared data association can be based in NFC module
Meeting (IrDA) technology, ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, electronic equipment 900 can be by one or more application specific integrated circuit (ASIC), number
Word signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided
It such as include the memory 904 of instruction, above-metioned instruction can be executed by the processor 920 of electronic equipment 900 to complete the above method.Example
Such as, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, soft
Disk and optical data storage devices etc..
The embodiment of the present application also provides a kind of non-transitorycomputer readable storage mediums, when in the storage medium
When instruction is executed by the processor of electronic equipment, so that electronic equipment is able to carry out any audio matter in above-described embodiment
The detection method of amount.
The embodiment of the present application also provides a kind of application product, the application product for executing at runtime
State the detection method of any audio quality in embodiment.
Those skilled in the art will readily occur to its of the application after considering specification and practicing application disclosed herein
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and including the undocumented common knowledge in the art of the application
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by above
Claim is pointed out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (10)
1. a kind of detection method of audio quality, which is characterized in that the described method includes:
Obtain multiple audios to be detected;
Extract the audio frequency characteristics of each audio to be detected, wherein the audio frequency characteristics are the mark audio to be detected
The feature of audio quality;
According to the audio frequency characteristics of each audio to be detected, the sequence knot of the audio quality of the multiple audio to be detected is determined
Fruit.
2. the method as described in claim 1, which is characterized in that the step of the audio frequency characteristics for extracting each audio to be detected
Suddenly, comprising:
Each audio to be detected is converted into frequency-region signal;
Calculate the effective spectrum bandwidth and/or frequency spectrum smoothness of the corresponding frequency-region signal of each audio to be detected;
The result being calculated is determined as to the audio frequency characteristics of each audio to be detected.
3. method according to claim 2, which is characterized in that described to calculate the corresponding frequency domain letter of each audio to be detected
Number effective spectrum bandwidth, comprising:
For the corresponding frequency-region signal of each audio to be detected, the energy of each frequency band of the frequency-region signal is calculated;
When the difference between the energy of the frequency band after the energy and n-th of frequency band of n-th of frequency band persistently reaches preset value, really
Fixed first frequency band to n-th of frequency band is effective band;
The effective spectrum bandwidth Y of the frequency-region signal is calculated according to formula Y=(fs/2)/K × n;
Wherein, fs is the sample rate of the audio to be detected, and K is the number of frequency bands of the frequency-region signal.
4. method according to claim 2, which is characterized in that described to calculate the corresponding frequency domain letter of each audio to be detected
Number frequency spectrum smoothness, comprising:
For the corresponding frequency-region signal of each audio to be detected, the energy of each frequency band of the frequency-region signal is calculated;
Calculate the covariance value between the energy of every two nearby frequency bands;
The average value for calculating the covariance value, the frequency spectrum smoothness as the corresponding frequency-region signal of the audio to be detected.
5. method as claimed in claim 4, which is characterized in that the association between the energy for calculating every two nearby frequency bands
Before the step of variance yields, the method also includes:
Time domain sliding average is carried out to the frequency-region signal according to formula s (t) *=s (t) × p+s (t-1) × (1-p);
Wherein, frequency-region signal corresponding to audio frame of s (t) * for t moment in the audio to be detected after time domain sliding average
Energy, s (t) be the audio to be detected in t moment audio frame corresponding to frequency-region signal energy, s (t-1) is described
The energy of frequency-region signal corresponding to the audio frame at t-1 moment in audio to be detected, p are default slide coefficient, p ∈ (0,1).
6. method according to claim 2, which is characterized in that the multiple audio to be detected is stereo audio;
Before the step of result that will be calculated is determined as the audio frequency characteristics of each audio to be detected, the side
Method further include:
The Stereo-width for calculating the corresponding frequency-region signal of each audio to be detected, obtains calculated result.
7. as the method according to claim 1 to 6, which is characterized in that the sound according to each audio to be detected
Frequency feature, the step of determining the ranking results of the audio quality of the multiple audio to be detected, comprising:
The audio frequency characteristics of each audio to be detected are normalized according to preset rules respectively, obtain corresponding numerical value;
Default weight is distributed for each numerical value, and calculates the corresponding weighted value of each audio to be detected;
According to the corresponding weighted value of each audio to be detected, the sequence of the audio quality of the multiple audio to be detected is determined
As a result.
8. a kind of detection device of audio quality, which is characterized in that described device includes:
Audio to be detected obtains module, is configured as obtaining multiple audios to be detected;
Audio feature extraction module is configured as extracting the audio frequency characteristics of each audio to be detected, wherein the audio is special
Sign is to identify the feature of the audio quality of the audio to be detected;
Ranking results determining module is configured as the audio frequency characteristics according to each audio to be detected, determine it is the multiple to
Detect the ranking results of the audio quality of audio.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein described
Processor, the communication interface, the memory complete mutual communication by communication bus;
The memory, for storing computer program;
The processor when for executing the program stored on the memory, realizes side as claimed in claim 1 to 7
Method step.
10. a kind of non-transitorycomputer readable storage medium, which is characterized in that when the instruction in the storage medium is by electronics
When the processor of equipment executes, so that the electronic equipment is able to carry out method and step as claimed in claim 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910284120.1A CN110033784B (en) | 2019-04-10 | 2019-04-10 | Audio quality detection method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910284120.1A CN110033784B (en) | 2019-04-10 | 2019-04-10 | Audio quality detection method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110033784A true CN110033784A (en) | 2019-07-19 |
CN110033784B CN110033784B (en) | 2020-12-25 |
Family
ID=67237816
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910284120.1A Active CN110033784B (en) | 2019-04-10 | 2019-04-10 | Audio quality detection method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110033784B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111653286A (en) * | 2019-12-31 | 2020-09-11 | 广州励丰文化科技股份有限公司 | Hot backup method of audio signal and audio equipment |
WO2021025622A1 (en) * | 2019-08-05 | 2021-02-11 | National University Of Singapore | System and method for assessing quality of a singing voice |
CN113345467A (en) * | 2021-05-19 | 2021-09-03 | 苏州奇梦者网络科技有限公司 | Method, device, medium and equipment for evaluating spoken language pronunciation |
CN115171729A (en) * | 2022-09-02 | 2022-10-11 | 北京达佳互联信息技术有限公司 | Audio quality determination method and device, electronic equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101411171A (en) * | 2006-01-31 | 2009-04-15 | 艾利森电话股份有限公司 | Non-intrusive signal quality assessment |
US20100189290A1 (en) * | 2009-01-29 | 2010-07-29 | Samsung Electronics Co. Ltd | Method and apparatus to evaluate quality of audio signal |
CN103918030A (en) * | 2011-09-29 | 2014-07-09 | 杜比国际公司 | High quality detection in fm stereo radio signals |
CN104391282A (en) * | 2014-11-24 | 2015-03-04 | 天津大学 | Method for improving imaging quality by spectrum correction |
CN104981866A (en) * | 2013-01-04 | 2015-10-14 | 华为技术有限公司 | Method for determining a stereo signal |
CN105989853A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Audio quality evaluation method and system |
CN109285556A (en) * | 2018-09-29 | 2019-01-29 | 百度在线网络技术(北京)有限公司 | Audio-frequency processing method, device, equipment and storage medium |
CN109545242A (en) * | 2018-12-07 | 2019-03-29 | 广州势必可赢网络科技有限公司 | A kind of audio data processing method, system, device and readable storage medium storing program for executing |
-
2019
- 2019-04-10 CN CN201910284120.1A patent/CN110033784B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101411171A (en) * | 2006-01-31 | 2009-04-15 | 艾利森电话股份有限公司 | Non-intrusive signal quality assessment |
US20100189290A1 (en) * | 2009-01-29 | 2010-07-29 | Samsung Electronics Co. Ltd | Method and apparatus to evaluate quality of audio signal |
CN103918030A (en) * | 2011-09-29 | 2014-07-09 | 杜比国际公司 | High quality detection in fm stereo radio signals |
CN104981866A (en) * | 2013-01-04 | 2015-10-14 | 华为技术有限公司 | Method for determining a stereo signal |
CN104391282A (en) * | 2014-11-24 | 2015-03-04 | 天津大学 | Method for improving imaging quality by spectrum correction |
CN105989853A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Audio quality evaluation method and system |
CN109285556A (en) * | 2018-09-29 | 2019-01-29 | 百度在线网络技术(北京)有限公司 | Audio-frequency processing method, device, equipment and storage medium |
CN109545242A (en) * | 2018-12-07 | 2019-03-29 | 广州势必可赢网络科技有限公司 | A kind of audio data processing method, system, device and readable storage medium storing program for executing |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021025622A1 (en) * | 2019-08-05 | 2021-02-11 | National University Of Singapore | System and method for assessing quality of a singing voice |
US11972774B2 (en) | 2019-08-05 | 2024-04-30 | National University Of Singapore | System and method for assessing quality of a singing voice |
CN111653286A (en) * | 2019-12-31 | 2020-09-11 | 广州励丰文化科技股份有限公司 | Hot backup method of audio signal and audio equipment |
CN113345467A (en) * | 2021-05-19 | 2021-09-03 | 苏州奇梦者网络科技有限公司 | Method, device, medium and equipment for evaluating spoken language pronunciation |
CN113345467B (en) * | 2021-05-19 | 2023-10-20 | 苏州奇梦者网络科技有限公司 | Spoken language pronunciation evaluation method, device, medium and equipment |
CN115171729A (en) * | 2022-09-02 | 2022-10-11 | 北京达佳互联信息技术有限公司 | Audio quality determination method and device, electronic equipment and storage medium |
CN115171729B (en) * | 2022-09-02 | 2023-01-20 | 北京达佳互联信息技术有限公司 | Audio quality determination method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110033784B (en) | 2020-12-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110033784A (en) | A kind of detection method of audio quality, device, electronic equipment and storage medium | |
CN111294704B (en) | Audio processing method, device and storage medium | |
CN107146631B (en) | Music identification method, note identification model establishment method, device and electronic equipment | |
CN109493852A (en) | A kind of evaluating method and device of speech recognition | |
US11284151B2 (en) | Loudness adjustment method and apparatus, and electronic device and storage medium | |
CN103944804A (en) | Contact recommending method and device | |
CN106250430A (en) | The sort method of smart machine list and device | |
CN104636453A (en) | Illegal user data identification method and device | |
CN103986999A (en) | Method, device and terminal equipment for detecting earphone impedance | |
CN109862421A (en) | A kind of video information recognition methods, device, electronic equipment and storage medium | |
CN114598970A (en) | Audio processing method and device, electronic equipment and storage medium | |
US11682412B2 (en) | Information processing method, electronic equipment, and storage medium | |
CN110490303A (en) | Super-network construction method, application method, device and medium | |
CN109375893A (en) | Audio frequency playing method and device | |
CN109729382A (en) | Volume processing method, device and server, the electronic equipment of audio, video data | |
CN106020766A (en) | Music playing method and device | |
CN108268667A (en) | Audio file clustering method and device | |
CN109524025B (en) | Singing scoring method and device, electronic equipment and storage medium | |
CN106020646A (en) | Media volume adjusting method and apparatus, and terminal device | |
CN109066865A (en) | The calibration method and terminal of battery capacity | |
CN109088920B (en) | Evaluation method, device and equipment of intelligent sound box and storage medium | |
CN105933775A (en) | Video playing progress adjusting method and device and terminal equipment | |
CN105549950A (en) | Sound production method and apparatus for portable electronic device | |
CN107172557B (en) | Method and device for detecting polarity of loudspeaker and receiver | |
CN111694539A (en) | Method, apparatus and medium for switching between earpiece and speaker |
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 |