Explosive sound detection method and device and storage medium

Abstract

The invention relates to a plosive detection method and device and a storage medium. The plosive sound detection method comprises the following steps: collecting an audio signal sent by audio equipment; and determining a plurality of spectral energies of the audio signal in a time domain. And performing normalization processing on the plurality of frequency spectrum energies to obtain a plurality of quantized values, the plurality of quantized values and the plurality of frequency spectrum energies having a one-to-one correspondence relationship. And performing short-time Fourier transform on the audio signal to obtain a plurality of frequency spectrum energy sums on a time sequence. And detecting the plosive sound existing in the audio signal sent by the audio equipment based on the plurality of frequency spectrum energies and the quantized values on the time sequence. According to the invention, detection of slight and even covered plosive sound can be realized.

Description

Technical Field

[0001] The present disclosure relates to the field of audio performance detection of speaker units, and in particular to a method, device and storage medium for detecting plosive sounds. Background Art

[0002] In the related art, there are many solutions based on soundcheck to realize the electroacoustic performance detection of speakers, and soundcheck itself is an accurate and powerful software-based electroacoustic and audio electronic measurement system, which can quickly measure parameters such as frequency response, phase, sensitivity, distortion, directivity, impedance and abnormal sound. However, the related technology cannot measure the short-term explosive sound noise emitted by the speaker during the sounding process, that is, POP sound, which affects the actual effect of the blast sound detection. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a plosive sound detection method, device and storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, a method for detecting plosive sound is provided, comprising:

[0005] The audio signal emitted by the audio device is collected, and multiple spectrum energies of the audio signal in the time domain are determined; the multiple spectrum energies are normalized to obtain multiple quantized values, and there is a one-to-one correspondence between the multiple quantized values ​​and the multiple spectrum energies; the audio signal is short-time Fourier transformed to obtain multiple spectrum energy sums in a time series; based on the multiple spectrum energy and quantized values ​​in the time series, plosive sounds present in the audio signal emitted by the audio device are detected.

[0006] In one implementation, the normalizing the multiple spectrum energies to obtain multiple quantized values ​​and the normalizing the time-frequency information of the audio signal to obtain multiple quantized values ​​include:

[0007] Based on the time-frequency information of the audio signal, a first confidence interval is determined according to a first confidence level, and the spectrum data energies located in the first confidence interval among the multiple spectrum data energies are determined; the mean of the spectrum energies of the spectrum data located in the first confidence interval is determined; a parameter value for normalizing the mean is determined, and the spectrum energies of the multiple spectrum data are normalized based on the parameter value to obtain multiple quantized values ​​corresponding to the spectrum energies of the multiple spectrum data, and there is a one-to-one correspondence between the multiple quantized values ​​and the multiple spectrum data.

[0008] In one implementation, the detecting, based on the multiple quantized values, the plosive sound present in the audio signal emitted by the audio device comprises:

[0009] Based on the multiple quantization values, determine a broad-spectrum noise quantization value corresponding to the broad-spectrum noise value, wherein the broad-spectrum noise is a steady-state noise emitted by an audio signal for testing; determine a difference between the multiple quantization values ​​and the broad-spectrum noise quantization value; and determine the spectral energy and characterized audio corresponding to the quantization value whose difference is greater than a plosive sound detection threshold as a plosive sound.

[0010] In one implementation, determining the broad-spectrum noise quantization value corresponding to the broad-spectrum noise value based on the multiple quantization values ​​includes:

[0011] A second confidence interval is determined according to a second confidence level, and a quantization value located in the second confidence interval among the multiple quantization values ​​is determined; and a maximum quantization value among the means of the quantization values ​​located in the second confidence interval is used as the broad-spectrum noise quantization value corresponding to the broad-spectrum noise value.

[0012] In one implementation, the spectrum data determines a plurality of spectrum energies of the audio signal in the time domain, including:

[0013] Performing a time-frequency domain transform on the audio signal to obtain spectral energy of the audio signal in the time domain; dividing the time domain into multiple time intervals according to a preset time interval, and summing the spectral energy in each of the multiple time intervals to obtain the sum of spectral energies corresponding to the multiple time intervals; and using the multiple spectral energy sums as multiple spectral energies of the audio signal in the time domain.

[0014] According to a second aspect of an embodiment of the present disclosure, there is provided a plosive sound detection device, comprising:

[0015] The collecting unit is used to collect the audio signal emitted by the audio device and determine multiple spectrum energies of the audio signal in the time domain; the processing unit is used to normalize the multiple spectrum energies to obtain multiple quantization values, and there is a one-to-one correspondence between the multiple quantization values ​​and the multiple spectrum energies; the detection unit is used to detect the plosive sound in the audio signal emitted by the audio device based on the multiple quantization values.

[0016] In one implementation, the processing unit performs normalization processing on the multiple spectrum energies to obtain multiple quantized values ​​in the following manner:

[0017] Determine a first confidence interval according to a first confidence level, and determine the spectrum energy in the first confidence interval among the multiple spectrum energies; determine the mean of the spectrum energy in the first confidence interval; determine a parameter value for normalizing the mean, and normalize the multiple spectrum energies based on the parameter value to obtain multiple quantized values ​​corresponding to the multiple spectrum energies.

[0018] In one implementation, the detection unit detects the plosive sound in the audio signal emitted by the audio device based on the multiple quantization values ​​in the following manner:

[0019] Based on the multiple quantization values, a broad-spectrum noise quantization value is determined, where the broad-spectrum noise is a steady-state noise emitted by an audio signal for testing; a difference between the multiple quantization values ​​and the broad-spectrum noise quantization value is determined; and audio represented by spectral energy corresponding to a quantization value whose difference is greater than a plosive sound detection threshold is determined as a plosive sound.

[0020] In one implementation, the detection unit determines the broad-spectrum noise quantization value based on the multiple quantization values ​​in the following manner:

[0021] A second confidence interval is determined according to a second confidence level, and a quantization value located in the second confidence interval among the multiple quantization values ​​is determined; and a maximum quantization value among the quantization values ​​located in the second confidence interval is used as a broad-spectrum noise quantization value.

[0022] In one implementation, the acquisition unit determines multiple spectrum energies of the audio signal in the time domain in the following manner:

[0023] Performing a time-frequency domain transform on the audio signal to obtain spectral energy of the audio signal in the time domain; dividing the time domain into multiple time intervals according to a preset time interval, and summing the spectral energy in each of the multiple time intervals to obtain the sum of spectral energies corresponding to the multiple time intervals; and using the multiple spectral energy sums as multiple spectral energies of the audio signal in the time domain.

[0024] According to a third aspect of an embodiment of the present disclosure, there is provided an audio control device, including:

[0025] A processor; a memory for storing processor executable instructions; wherein the processor is configured to: execute the function control method described in the first aspect or any one of the embodiments of the first aspect.

[0026] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided, in which instructions are stored. When the instructions in the storage medium are executed by a processor of a terminal, the terminal can perform the method described in the first aspect or any one of the implementations of the first aspect.

[0027] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: the test device collects the audio signal emitted by the audio device, determines the spectrum energy of the collected audio signal in the time domain, performs normalization processing, obtains a quantized value to detect the plosive sound in the audio signal, thereby achieving the test of any audio signal and the detection of slight plosive sounds or even masked plosive sounds.

[0028] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0030] Figure 1 is a schematic diagram of a plosive sound testing environment according to an exemplary embodiment.

[0031] Figure 2 It is a schematic diagram showing a time-frequency spectrum including a POP sound according to an exemplary embodiment.

[0032] Figure 3 The figure is a flow chart of a method for detecting plosive sound according to an exemplary embodiment.

[0033] Figure 4 The figure is a flow chart of a spectrum energy normalization method according to an exemplary embodiment.

[0034] Figure 5 The figure is a flow chart of a method for detecting plosive sound according to an exemplary embodiment.

[0035] Figure 6 The figure is a flow chart showing a method for determining a broad-spectrum noise quantization value according to an exemplary embodiment.

[0036] Figure 7 The figure is a flow chart of a spectrum energy acquisition method according to an exemplary embodiment.

[0037] Figure 8 It is a schematic diagram showing a correspondence between time and spectrum information according to an exemplary embodiment.

[0038] Fig. 9 The figure is a flowchart of a method for detecting a plosive sound according to an exemplary embodiment.

[0039] Fig.10 The diagram is a schematic diagram showing a time and a difference between a plurality of quantization values ​​and a broad-spectrum noise quantization value according to an exemplary embodiment.

[0040] Fig.11 is a block diagram of a plosive sound detection device according to an exemplary embodiment.

[0041] Fig.12 It is a block diagram of a device for detecting plosive sounds according to an exemplary embodiment. DETAILED DESCRIPTION

[0042] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

[0043] In the accompanying drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limitations on the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present disclosure. The embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings.

[0044] The plosive sound detection method provided in the embodiment of the present disclosure can be applied to the application scenario of testing the audio performance of the speaker unit. By introducing audio detection including confidence interval calculation, it is possible to accurately determine whether there is a plosive sound in the sound-emitting audio device. It should be understood that the plosive sound can also be expressed as a POP sound. The present disclosure solves the problem that the detection conditions for POP sound are harsh, and only the POP sound generated during the steady-state signal jump process can be calculated. At the same time, the POP sound requirements are very obvious, and the POP sound with smaller energy or the POP generated by other reasons such as distortion cannot be judged, especially the pop sound that is masked by the original signal in the time domain and basically cannot be judged.

[0045] Figure 1 is a schematic diagram of a plosive sound testing environment according to an exemplary embodiment.

[0046] In the following embodiments of the present disclosure, the detection device 100 is a computer, the signal conversion device 101 is a sound card, the sound-generating audio device 102 is a True Wireless Stereo (TWS) headset, and the sound receiving device 103 is a microphone as an example to illustrate the audio control method involved in the above embodiments. First, the TWS headset continuously uses the Bluetooth audio transmission model protocol (Advanced Audio Distribution Profile, A2DP) to play a specific signal, which is then recorded by the artificial ear and transmitted back to the computer through the sound card. The computer locally calculates and analyzes the POP sound of the recorded signal.

[0047] Figure 2 It is a schematic diagram showing a time-frequency spectrum including a POP sound according to an exemplary embodiment.

[0048] In the related art, only the energy contained in adjacent frames of the audio signal is calculated by difference, and only the POP sound generated during the steady-state signal jump process can be calculated, for example Figure 2 The identifiable POP sound signals marked in the figure cannot identify POP sounds with less energy or caused by other reasons such as distortion, especially POP sounds that are masked by the original signal in the time domain, such as Figure 2 The difficult-to-identify POP sound signal marked in the figure.

[0049] In view of this, an embodiment of the present disclosure provides a method for detecting plosive sounds. In the method for detecting plosive sounds provided by the present disclosure, the acquired audio data is quantized by normalization, broad-spectrum noise removal, and the like, and compared with a detection threshold, and audio greater than the detection threshold is determined as a plosive sound. This enables detection of smaller or even masked POP sounds.

[0050] Figure 3 is a flow chart of a method for detecting plosive sound according to an exemplary embodiment. Figure 3 As shown, the plosive sound control method includes the following steps.

[0051] In step S11, an audio signal emitted by an audio device is collected, and multiple spectrum energies of the audio signal in the time domain are determined.

[0052] In the embodiment of the present disclosure, the audio signal at least includes a fixed-frequency signal for testing and a plosive sound signal emitted by the audio device. Before determining multiple spectrum energies of the audio signal in the time domain, the audio signal can also be filtered to filter out ambient sound that may interfere with the detection.

[0053] In the disclosed embodiment, during the test, the audio device needs to play a steady-state signal, such as white noise or other steady-state noise or a fixed-frequency signal, to avoid test errors caused by sudden changes in the test audio signal.

[0054] In step S12, a normalization process is performed on the multiple spectrum energies to obtain multiple quantization values, and there is a one-to-one correspondence between the multiple quantization values ​​and the multiple spectrum energies.

[0055] In the embodiments of the present disclosure, the spectral energy of the audio data includes the time domain characteristics and frequency domain characteristics of the audio data. For example, the time-frequency spectrum of the audio data is obtained by short-time Fourier transform. It should be understood that the method for obtaining the spectral energy may be other methods besides short-time Fourier transform, which is not limited in the present disclosure.

[0056] In the disclosed embodiment, the spectrum energy data is normalized to standardize the data of each test and unify the data obtained from different batches and different conditions. Normalization maps the data into a certain range so that the data has the same numerical range to facilitate analysis and comparison.

[0057] In step S13, based on a plurality of quantized values ​​in time series, a plosive sound present in the audio signal emitted by the audio device is detected.

[0058] In the disclosed embodiment, the plosive sound contained in the audio signal is obtained by detecting the normalized quantized value, wherein the normalized quantized value of the plosive sound is significantly different from the quantized value not containing the plosive sound.

[0059] In the embodiment of the present disclosure, normalization of the spectrum energy further includes normalization through the mean of the confidence interval.

[0060] Figure 4 is a flow chart of a spectrum energy normalization method according to an exemplary embodiment. Figure 4 As shown, the spectrum energy normalization method includes the following steps.

[0061] In step S21, a first confidence interval is determined according to a first confidence level, and spectrum data energies in the first confidence interval among a plurality of spectrum data energies are determined.

[0062] In the embodiment of the present disclosure, the first confidence interval represents a range of spectrum energy, and the first confidence level is used to represent the probability of randomly selecting a spectrum energy and it being within the range represented by the corresponding first confidence interval. Therefore, if the first confidence level is a first confidence interval of n, then there are n×m spectrum energies in the total number of spectrum energies m that are located in the first confidence interval. For example, assuming that there are 10 spectrum energies, and the first confidence level corresponding to the first confidence interval is 0.9, then 9 spectrum energies in the 10 time-frequency information are located in the first confidence interval.

[0063] In step S22, the mean value of the spectrum energy within the first confidence interval is determined.

[0064] By determining the mean value of the spectrum energy in the first confidence interval, a standardization operation on the spectrum energy data is achieved.

[0065] In step S23, a parameter value for normalizing the mean is determined, and a plurality of spectrum energies are normalized based on the parameter value to obtain a plurality of quantized values ​​corresponding to the plurality of spectrum energies.

[0066] In the embodiment of the present disclosure, if the mean of the spectrum data of the first confidence interval multiplied by a parameter value is equal to 1, then the parameter value is the reciprocal of the mean of the spectrum data of the first confidence interval. The spectrum data is normalized by multiplying all the spectrum data by the parameter value. Normalization realizes the standardization of the spectrum data so that it can be comparable with other normalized audio data.

[0067] In the embodiment of the present disclosure, the normalization processing of the spectrum energy is realized. However, based on the obtained multiple quantization values, it is necessary to remove the broad spectrum noise and compare with the detection threshold to realize the detection of the plosive sound in the audio signal.

[0068] Figure 5 is a flow chart of a method for detecting plosive sound according to an exemplary embodiment. Figure 5 As shown, the plosive sound detection method includes the following steps.

[0069] In step S31 , a broad spectrum noise quantization value is determined based on a plurality of quantization values, where the broad spectrum noise is a steady-state noise emitted by an audio signal for testing.

[0070] In the disclosed embodiment, the broad-spectrum noise quantization value is used to represent the quantization value of the spectrum energy contained in the audio signal emitted by the audio device during the test. By determining the broad-spectrum noise quantization value, the dynamic adjustment process of the broad-spectrum noise after normalization can be reflected.

[0071] In step S32, the difference between the plurality of quantized values ​​and the broad spectrum noise quantized value is determined.

[0072] In the embodiment of the present disclosure, the normalized data is used to subtract the broad spectrum noise quantization value, so as to obtain the quantization value of the spectrum energy after removing most of the broad spectrum noise. By determining the difference between the multiple quantization values ​​and the broad spectrum noise quantization value, the quantization value data of the spectrum energy mainly containing the plosive sound can be obtained.

[0073] In step S33, the audio represented by the spectrum energy corresponding to the quantized value whose difference is greater than the plosive sound detection threshold is determined to be a plosive sound.

[0074] In the disclosed embodiment, by comparing the difference with the plosive detection threshold, if it is greater than the plosive detection threshold, it is determined to be a plosive; if it is less than the plosive detection threshold, it is determined to be a non-plosive. It should be understood that the plosive detection threshold can be adjusted by the user according to subjective feelings. By comparing the difference with the plosive detection threshold, accurate detection of plosives can be achieved.

[0075] In the embodiment of the present disclosure, the detection of plosives is achieved by using the audio represented by the spectral energy corresponding to the quantized value whose difference is greater than the plosive detection threshold. However, the detection of plosives requires the specific confirmation of the quantized value of the broad spectrum noise.

[0076] Figure 6 is a flow chart showing a method for determining a broad spectrum noise quantization value according to an exemplary embodiment. Figure 6 As shown, the method for determining the broad-spectrum noise quantization value includes the following steps.

[0077] In step S41, a second confidence interval is determined according to a second confidence level, and a quantized value in the plurality of quantized values ​​that is in the second confidence interval is determined.

[0078] In the disclosed embodiment, the second confidence interval is used to represent a range of a quantized value, and the second confidence level is used to represent the probability that a quantized value is randomly selected and that the quantized value is within the range represented by the second confidence interval. The quantized value within the second confidence interval can be determined by the second confidence level and the second confidence interval.

[0079] In step S42, the maximum quantized value among the quantized values ​​in the second confidence interval is used as the broad-spectrum noise quantized value.

[0080] In the disclosed embodiment, the broad-spectrum noise quantization value is used to represent the quantization value of the audio energy and emitted by the test steady-state noise in the audio signal. The broad-spectrum noise quantization value is used to represent the magnitude of the broad-spectrum noise contained in the quantization value corresponding to each time, and the detection threshold is dynamically adjusted.

[0081] In the disclosed embodiment, by setting the second confidence interval and quantizing the broad spectrum noise, it is achieved that the spectrum energy can be determined by the time domain information and the frequency domain information corresponding to each time unit.

[0082] Figure 7 is a flow chart of a spectrum energy acquisition method according to an exemplary embodiment. Figure 7 As shown, the spectrum energy acquisition method includes the following steps.

[0083] In step S51, the audio signal is transformed in the time domain to obtain the spectrum energy of the audio signal in the time domain.

[0084] In the disclosed embodiment, the time domain and frequency domain characteristics of the audio signal can be obtained through time-frequency domain transformation, so as to obtain the frequency domain energy information of the audio signal in the time domain. For example, the sampling rate and window width are used to perform time-frequency domain transformation on the audio signal through STFT. Among them, the sampling rate defines the number of samples extracted from the continuous signal and composed of the discrete signal per unit time, and the window width is used to control the accuracy of the Fourier transform. The corresponding relationship between time and spectrum information can be obtained through the time domain information, sampling rate and window width of the audio signal.

[0085] In step S52, the time domain is divided into multiple time intervals according to a preset time interval, and the spectrum energy in each of the multiple time intervals is summed to obtain the spectrum energy sum corresponding to the multiple time intervals.

[0086] In the embodiment of the present disclosure, the preset time interval can be determined by a method of time-frequency domain transformation. For example, if STFT is used to transform the audio signal in the time-frequency domain, the time interval is determined by the sampling rate and the window width in the STFT. The spectrum energy in each interval is obtained by Fourier transforming the time domain information in the interval.

[0087] In step S53, the sum of the multiple spectrum energies is used as the multiple spectrum energies of the audio signal in the time domain.

[0088] In the disclosed embodiment, since the spectrum energy contained in the plosive sound is much greater than the spectrum energy contained in the steady-state noise, the spectrum energy value contained in each time correspondence is summed up, so that the spectrum energy value of the plosive sound can be more prominent.

[0089] Figure 8 The diagram is a schematic diagram showing a corresponding relationship between time and spectrum energy according to an exemplary embodiment.

[0090] The correspondence between time and spectrum information in this exemplary embodiment is as follows: Figure 8 As shown in the figure, assuming that the signal is sampled at 48000 Hz during recording, the time-frequency domain diagram is obtained by short-time Fourier transform (STFT), and the window width is 256. The unit of time in the obtained time-frequency domain diagram is the quotient of 48000 and 256. The ordinate in the time-frequency domain diagram represents the frequency, and the light and dark represent the energy contained in the frequency. If all the frequency energies corresponding to each time interval are summed, the corresponding relationship between time and the sum of spectrum energy can be obtained.

[0091] Fig. 9 The figure is a flowchart of a method for detecting a plosive sound according to an exemplary embodiment.

[0092] The plosive sound detection process of this exemplary embodiment is as follows Fig. 9As shown, it is assumed that the audio device is a TWS headset. It should be understood that the mean of the spectral energy of the first confidence interval can also be called a broad-spectrum benchmark, and the broad-spectrum noise quantization value can also be called the broad-spectrum noise basis. First of all, a steady-state signal needs to be played during the test. Considering that any signal can be transmitted in the A2DP mode of the TWS headset, the headset can play steady-state noise such as white noise or a fixed-frequency signal during recording, and the signal is recorded at 48,000 samples. The signal is collected by a microphone and a sound card. The test device processes the signal through STFT to obtain time-frequency domain information. The window width can be 256. Since the POP sound does not end instantly, the window width value does not need to be too high. From the sampling rate and window width, it can be concluded that the time unit is 256 / 48000 seconds, which is about 5 milliseconds. After that, the frequency band data is filtered out, and the spectrum data above 1000hz is intercepted to reduce the additional impact of ambient sound. Spectral energy summation: sum the spectrum data along the time axis to obtain the spectrum energy and data that changes with the time axis. The spectrum energy sum is the sum of all spectrum energies within the current 5 milliseconds. Since the energy contained in POP sound is much greater than the energy of steady-state noise, summing the spectrum energy can better reflect the difference between POP sound and steady-state noise. Therefore, the spectrum energy sum can prove POP sound. For example, the spectrum energy detected within a certain 5 milliseconds is 10dB, 20dB, and 15dB, respectively, while the spectrum energy detected within another 5 milliseconds is 30dB, 15dB, and 40dB, respectively. The energy sum can clearly show that the latter has POP sound. Normalize the spectrum data, calculate the confidence interval of the curve according to the confidence level of 0.9, and intercept the average value of the confidence interval data as the broad spectrum benchmark. Modify all data sizes by normalizing to 1 based on the broad spectrum benchmark. To remove broad-spectrum noise, calculate the normalized curve confidence interval according to the confidence level of 0.8, intercept the maximum value of the confidence interval data as the basis of broad-spectrum noise, use the normalized data to subtract the basis of broad-spectrum noise, and obtain the corresponding relationship between time and the difference between the broad-spectrum reference and the basis of broad-spectrum noise. By removing the broad-spectrum noise operation, the broad-spectrum noise in the audio can be dynamically removed. At the same time, a threshold is set, and when the difference exceeds the threshold design point, it is a POP sound.

[0093] Fig.10 The diagram is a schematic diagram showing a time and a difference between a plurality of quantization values ​​and a broad-spectrum noise quantization value according to an exemplary embodiment.

[0094] The difference between the exemplary time and quantized value and the broad spectrum noise quantized value is as follows Fig.10 As shown in the figure, the horizontal axis represents time, the vertical axis represents the difference between the quantized value and the broad-spectrum noise quantized value, delta represents the relationship curve between time and the difference between the quantized value and the broad-spectrum noise quantized value, and limit represents the pop sound detection threshold. When delta exceeds limit, it means that a pop sound appears at the time corresponding to the delta exceeding limit.

[0095] In the disclosed embodiment, by using the correspondence between time and spectrum energy, the spectrum energy is normalized and the broad-spectrum noise is removed, thereby achieving accurate judgment of the existence of plosive sounds and avoiding missed detection of plosive sounds emitted by audio equipment.

[0096] Based on the same concept, an embodiment of the present disclosure also provides a plosive sound detection device.

[0097] It is understandable that, in order to realize the above functions, the plosive sound detection device provided in the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function. In combination with the units and algorithm steps of each example disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0098] Fig.11 is a block diagram of a plosive sound detection device 200 according to an exemplary embodiment. Fig.11 The device includes a collection unit 201, a processing unit 202 and a detection unit 203.

[0099] The collection unit 201 is used to collect audio signals emitted by an audio device and determine multiple spectrum energies of the audio signals in the time domain.

[0100] The processing unit 202 is used to perform normalization processing on the multiple spectrum energies to obtain multiple quantization values, and there is a one-to-one correspondence between the multiple quantization values ​​and the multiple spectrum energies.

[0101] The detection unit 203 is configured to detect, based on the multiple quantization values, the plosive sound present in the audio signal emitted by the audio device.

[0102] In one embodiment, the processing unit 202 normalizes the multiple spectral energies to obtain multiple quantized values ​​in the following manner: determining a first confidence interval according to a first confidence level, and determining the spectral energies in the multiple spectral energies that are located in the first confidence interval; determining the mean of the spectral energies located in the first confidence interval; determining a parameter value for normalizing the mean, and normalizing the multiple spectral energies based on the parameter value to obtain multiple quantized values ​​corresponding to the multiple spectral energies.

[0103] In one embodiment, the detection unit 203 detects plosive sounds in an audio signal emitted by an audio device based on multiple quantization values ​​in the following manner: determining a quantization value of a broad-spectrum noise based on the multiple quantization values, where the broad-spectrum noise is a steady-state noise emitted by the audio signal for testing; determining a difference between the multiple quantization values ​​and the quantization value of the broad-spectrum noise; and determining the audio represented by the spectral energy corresponding to the quantization value whose difference is greater than a plosive sound detection threshold as a plosive sound.

[0104] In one embodiment, the detection unit 203 determines the broad-spectrum noise quantization value based on multiple quantization values ​​in the following manner: determining a second confidence interval according to a second confidence level, determining a quantization value in the multiple quantization values ​​that is in the second confidence interval; and using the maximum quantization value in the quantization values ​​in the second confidence interval as the broad-spectrum noise quantization value.

[0105] In one embodiment, the acquisition unit 201 determines multiple spectrum energies of the audio signal in the time domain in the following manner: based on the time domain signal, sampling rate and window width of the audio signal, obtain the correspondence between time and spectrum information, where the time in the correspondence is determined by the sampling rate and the window width; based on the correspondence between time and spectrum information, determine the energy contained in all frequencies corresponding to each moment; the energy contained in all frequencies corresponding to each moment, and obtain the energy sum corresponding to each moment.

[0106] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0107] Fig.12 is a block diagram of an apparatus 300 for detecting plosive sounds according to an exemplary embodiment. For example, the apparatus 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0108] Reference Fig.12 , the device 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .

[0109] The processing component 302 generally controls the overall operation of the device 300, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 302 may include one or more modules to facilitate the interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate the interaction between the multimedia component 308 and the processing component 302.

[0110] The memory 304 is configured to store various types of data to support operations on the device 300. Examples of such data include instructions for any application or method operating on the device 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0111] The power component 306 provides power to the various components of the device 300. The power component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 300.

[0112] The multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0113] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC), and when the device 300 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 304 or sent via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0114] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0115] The sensor assembly 314 includes one or more sensors for providing various aspects of the status assessment of the device 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the device 300, the sensor assembly 314 can also detect the position change of the device 300 or a component of the device 300, the presence or absence of user contact with the device 300, the orientation or acceleration / deceleration of the device 300, and the temperature change of the device 300. The sensor assembly 314 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 can also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0116] The communication component 316 is configured to facilitate wired or wireless communication between the device 300 and other devices. The device 300 can access a wireless network based on a communication standard, such as WiFi, 3G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0117] In an exemplary embodiment, the apparatus 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0118] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, and the instructions can be executed by the processor 320 of the device 300 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0119] It is to be understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0120] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0121] It will be further understood that the terms “center”, “longitudinal”, “lateral”, “lower”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present embodiment and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation.

[0122] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0123] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0124] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modifications, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0125] It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. A method for detecting plosive sound, characterized in that: include: Collecting an audio signal emitted by an audio device, and determining multiple spectrum energies of the audio signal in the time domain; Normalizing the multiple spectrum energies to obtain multiple quantized values, where the multiple quantized values ​​have a one-to-one correspondence with the multiple spectrum energies; Based on the multiple quantized values, a plosive sound present in the audio signal emitted by the audio device is detected.

2. The method according to claim 1, characterized in that The normalizing process is performed on the multiple spectrum energies to obtain multiple quantized values, including: Determine a first confidence interval according to the first confidence level, and determine the spectrum energy in the plurality of spectrum energies that is located in the first confidence interval; determining a mean value of spectral energy within a first confidence interval; A parameter value for normalizing the mean is determined, and the multiple spectrum energies are normalized based on the parameter value to obtain multiple quantized values ​​corresponding to the multiple spectrum energies.

3. The method according to claim 1 or 2, characterized in that: The detecting, based on the multiple quantized values, the plosive sound present in the audio signal emitted by the audio device comprises: Determine a broad spectrum noise quantization value based on the multiple quantization values, wherein the broad spectrum noise is a steady-state noise emitted by the audio signal for testing; determining a difference between the plurality of quantized values ​​and the broad spectrum noise quantized value; The audio represented by the spectrum energy corresponding to the quantized value whose difference is greater than the plosive sound detection threshold is determined as a plosive sound.

4. The method according to claim 3, characterized in that The step of determining a broad spectrum noise quantization value based on the multiple quantization values ​​comprises: Determine a second confidence interval according to a second confidence level, and determine a quantized value in the plurality of quantized values ​​that is in the second confidence interval; The maximum quantized value among the quantized values ​​in the second confidence interval is used as the broad-spectrum noise quantized value.

5. The method according to claim 1, characterized in that The determining a plurality of spectrum energies of the audio signal in the time domain comprises: Performing a time-frequency domain transformation on the audio signal to obtain the spectrum energy of the audio signal in the time domain; Dividing the time domain into multiple time intervals according to a preset time interval, and summing the spectrum energy in each of the multiple time intervals to obtain the spectrum energy sum corresponding to the multiple time intervals; The sum of the multiple spectrum energies is used as multiple spectrum energies of the audio signal in the time domain.

6. A plosive sound detection device, characterized in that: include: A collection unit, used for collecting audio signals emitted by an audio device, and determining multiple spectrum energies of the audio signals in the time domain; A processing unit, configured to perform normalization processing on the multiple spectrum energies to obtain multiple quantization values, wherein the multiple quantization values ​​and the multiple spectrum energies have a one-to-one correspondence; The detection unit is used to detect the plosive sound existing in the audio signal emitted by the audio device based on the multiple quantization values.

7. The device according to claim 6, characterized in that The processing unit performs normalization processing on the multiple spectrum energies to obtain multiple quantized values ​​in the following manner: Determine a first confidence interval according to the first confidence level, and determine the spectrum energy in the plurality of spectrum energies that is located in the first confidence interval; determining a mean value of spectral energy within a first confidence interval; A parameter value for normalizing the mean is determined, and the multiple spectrum energies are normalized based on the parameter value to obtain multiple quantized values ​​corresponding to the multiple spectrum energies.

8. The device according to claim 6 or 7, characterized in that The detection unit detects the plosive sound in the audio signal emitted by the audio device based on the multiple quantization values ​​in the following manner: Determine a broad spectrum noise quantization value based on the multiple quantization values, wherein the broad spectrum noise is a steady-state noise emitted by the audio signal for testing; determining a difference between the plurality of quantized values ​​and the broad spectrum noise quantized value; The audio represented by the spectrum energy corresponding to the quantized value whose difference is greater than the plosive sound detection threshold is determined as a plosive sound.

9. An audio control device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: execute the method described in any one of claims 1 to 5.

10. A storage medium, characterized in that: The storage medium stores instructions, and when the instructions in the storage medium are executed by a processor of the terminal, the terminal is enabled to execute the method according to any one of claims 1 to 5.

Images