CN113870885B - Bluetooth audio squeal detection and suppression method, device, medium, and apparatus - Google Patents
Bluetooth audio squeal detection and suppression method, device, medium, and apparatus Download PDFInfo
- Publication number
- CN113870885B CN113870885B CN202111456960.5A CN202111456960A CN113870885B CN 113870885 B CN113870885 B CN 113870885B CN 202111456960 A CN202111456960 A CN 202111456960A CN 113870885 B CN113870885 B CN 113870885B
- Authority
- CN
- China
- Prior art keywords
- audio
- howling
- current
- pitch
- bluetooth
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L2021/02082—Noise filtering the noise being echo, reverberation of the speech
Abstract
The application discloses a Bluetooth audio squeal detection and suppression method, a device, a medium and equipment, belonging to the technical field of audio coding and decoding, wherein the method comprises the following steps: in the process of coding the Bluetooth audio by the long-term post filter, acquiring a fundamental tone existence mark of a current audio frame; if the fundamental tone existence mark indicates that the current audio frame has fundamental tone, determining the current fundamental tone delay of the current audio frame in the processing process of the long-term post filter, and calculating to obtain a corresponding current autocorrelation value; judging whether the pitch delays corresponding to a first preset number of previous continuous audio frames including the current audio frame are the same or not; if the first preset number of audio frames are the same, judging whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase; if yes, the Bluetooth audio has howling, and howling suppression is started. The method and the device have the advantages that the existing information of the long-term post-filter in the audio coding process is utilized to detect and inhibit the audio howling, the complexity is low, the calculation amount is small, and the power consumption is low.
Description
Technical Field
The present application relates to the field of audio encoding and decoding technologies, and in particular, to a method, an apparatus, a medium, and a device for detecting and suppressing a bluetooth audio squeal.
Background
In the mainstream bluetooth audio encoder at present, the "subband coding technology (SBC)" is used most widely, and all bluetooth audio devices must support, but the tone quality is general; the advanced audio coding technology (AAC-LC) has good tone quality and wide application, is supported by a plurality of mainstream mobile phones, but has larger memory occupation and high operation complexity, a plurality of Bluetooth devices are based on an embedded platform, the battery capacity is limited, the operation capability of a processor is poorer, and the memory is limited; the high-pass Bluetooth audio coding technology (aptX series) and the Sony Bluetooth audio coding technology (LDAC) have good sound quality but high code rate, are respectively unique high-pass and Sony technologies and are closed. For the above reasons, the bluetooth international union has introduced LC3 by a number of manufacturers, which has the advantages of low delay, high sound quality and coding gain, and no special interest in the bluetooth field, and is receiving attention from a number of manufacturers.
There are many typical application scenarios for bluetooth audio, such as bluetooth-based karaoke, bluetooth-based wireless microphone/speaker, etc., which can be abstracted from the data stream as shown in fig. 1, and it can be seen that there is a feedback path (as indicated by the dashed arrow) from the speaker to the microphone, and this closed loop state makes it possible for the system to tend to be unstable when the gain is increased to a certain state, and to generate self-excited oscillation and generate howling, in order to detect and further avoid howling, in the prior art, a module can be inserted in the path of audio processing, such as before the audio encoder or after the audio decoder, to detect the generation of howling. In the existing scheme, the frequency domain-based method is relatively accurate and widely used, but the method needs to complete time-frequency transformation and detect howling based on the characteristics obtained from the frequency domain, so that the calculation amount is relatively large.
Disclosure of Invention
The method, the device, the medium and the equipment for detecting and suppressing the Bluetooth audio howling are provided by the application aiming at the problems that in the prior art, a howling detection module needs to be inserted into an audio processing path, the processing process is complex, the power consumption is high, the calculation amount is large, and the method is not suitable for being applied to low-power Bluetooth.
In one embodiment of the present application, a method for detecting and suppressing audio howling based on bluetooth is provided, which includes: in the process of coding the Bluetooth audio by the long-term post filter, acquiring a fundamental tone existence mark of a current audio frame; if the fundamental tone existence mark indicates that the current audio frame has fundamental tone, determining the current fundamental tone delay of the current audio frame in the processing process of the long-term post filter, and calculating to obtain a corresponding current autocorrelation value; judging whether the pitch delays corresponding to a first preset number of previous continuous audio frames including the current audio frame are the same or not; if the first preset number of audio frames are the same, judging whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase; if yes, the Bluetooth audio has howling, and howling suppression is started.
Optionally, if the pitch existence flag indicates that the pitch exists in the current audio frame, the method includes: judging whether the value of the pitch existence flag is a preset value or not; if yes, the fact that the current audio frame has fundamental tones is shown, and if not, the fact that the current audio frame does not have the fundamental tones is shown.
Optionally, the calculating to obtain the corresponding current autocorrelation value includes: and in the standard process of calculating the corresponding normalized autocorrelation value according to the current pitch delay, extracting an intermediate calculation value in the standard process as the current autocorrelation value.
Optionally, the determining whether the autocorrelation values respectively corresponding to the first preset number of audio frames continuously increase includes: in the first preset number of audio frames, if the autocorrelation values of the following audio frames are not less than the autocorrelation values of the preceding audio frames, the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase.
Optionally, performing howling suppression, including: in the process of coding the Bluetooth audio by the long-term post-filter, the howling suppression is respectively carried out on a second preset number of post-continuous audio frames including the current audio frame, and the method comprises the following steps: and in the processing process of the long-term post-filter of the second preset number of audio frames, setting the action parameters of the long-term post-filter to be preset values, and writing the preset values into the coded code stream, so that the subsequent decoder does not perform frequency enhancement processing when decoding the second preset number of audio frames.
Optionally, the howling suppression is performed on the second preset number of subsequent consecutive audio frames including the current audio frame, and the method further includes: determining a pitch frequency according to a current pitch delay corresponding to a current audio frame; and setting a notch filter according to the fundamental tone frequency, and filtering a second preset number of audio frames.
In one aspect of the present application, a bluetooth audio howling detection and suppression apparatus is provided, including: the squeal detection module detects whether squeal exists in the Bluetooth audio, and the squeal detection module comprises: the pitch mark detection module is used for acquiring a pitch mark of a current audio frame in the process of coding the Bluetooth audio by the long-term post filter; the autocorrelation value acquisition module is used for determining the current pitch delay of the current audio frame in the processing process of the long-term post filter when the pitch existence flag indicates that the current audio frame has pitch, and calculating to obtain the corresponding current autocorrelation value; the judging module is used for judging whether the pitch delays corresponding to a first preset number of previous continuous audio frames including the current audio frame are the same or not; if the first preset number of audio frames are the same, judging whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase; if yes, the Bluetooth audio has howling; and the howling suppression module is used for carrying out howling suppression on the Bluetooth audio with the howling.
Optionally, the howling suppression module includes a notch filter, and filters the bluetooth audio that generates howling through the notch filter to perform howling suppression.
In one aspect of the present application, a computer-readable storage medium is provided, which stores computer instructions, wherein the computer instructions are operable to execute the bluetooth audio howling detection and suppression method of any one of the first aspect.
In one aspect of the present application, a computer device is provided, which includes a processor and a memory, where the memory stores computer instructions, wherein: the processor operates the computer instructions to perform the bluetooth audio howling detection and suppression method of any of scheme one.
The beneficial effect of this application is: the Bluetooth audio howling detection and inhibition method, the Bluetooth audio howling detection and inhibition device, the Bluetooth audio howling detection and inhibition medium and the Bluetooth audio howling detection and inhibition equipment perform detection and inhibition on audio howling by fully utilizing the existing information of the long-term post-filter in the audio coding process, are low in complexity, small in calculation amount and low in power consumption, and are suitable for being applied to low-power Bluetooth.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 shows a schematic diagram of a bluetooth audio howling generation process;
fig. 2 illustrates one embodiment of the bluetooth audio howling detection and suppression method of the present application;
fig. 3 shows an example of the bluetooth audio howling detection and suppression method of the present application;
fig. 4 shows an example of the bluetooth audio howling detection and suppression method of the present application;
fig. 5 shows an embodiment of the bluetooth audio howling detection and suppression apparatus of the present application;
fig. 6 shows an example of the application of the bluetooth audio howling detection and suppression apparatus at the encoder side according to the present application;
fig. 7 shows a schematic diagram of an audio signal with howling;
fig. 8 shows a schematic diagram of audio detection corresponding to fig. 7.
With the above figures, there are shown specific embodiments of the present application, which will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises" and "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a product or apparatus that comprises a list of steps or elements is not necessarily limited to those elements explicitly listed, but may include other elements not expressly listed or inherent to such product or apparatus.
In order to detect to further avoid howling, in the prior art, a howling detection module may be inserted in the path of audio processing, for example, before the audio encoder or after the audio decoder, to detect the generation of howling and further avoid howling. The existing scheme is based on a frequency domain method for detection, is relatively accurate and wide in use, but needs to complete time-frequency transformation and detect howling based on the characteristics obtained from the frequency domain, and has the disadvantages of large calculation amount, large power consumption and extremely high calculation complexity. In the application process of the bluetooth low energy, the bluetooth low energy has a high limit on the power consumption, so the howling detection and suppression method has a great disadvantage when being applied to the bluetooth low energy.
Aiming at the problems, the existing information generated in the encoding process is directly extracted to detect the audio squeal in the audio encoding process, and an additional squeal detection module is not specially arranged, so that the squeal detection process is simple and efficient, extra calculation amount is not required to be added, the power consumption is saved, and the method is more suitable for the low-power Bluetooth field.
Therefore, the application provides a method, a device, a medium and equipment for detecting and suppressing the Bluetooth audio squeal. The Bluetooth audio howling detection and suppression method comprises the following steps: in the process of coding the Bluetooth audio by the long-term post filter, acquiring a fundamental tone existence mark of a current audio frame; if the fundamental tone existence mark indicates that the current audio frame has fundamental tone, determining the current fundamental tone delay of the current audio frame in the processing process of the long-term post filter, and calculating to obtain a corresponding current autocorrelation value; judging whether the pitch delays corresponding to a first preset number of previous continuous audio frames including the current audio frame are the same or not; if the first preset number of audio frames are the same, judging whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase; if yes, the Bluetooth audio has howling, and howling suppression is started.
The method for detecting and inhibiting the Bluetooth audio squeal obtains a fundamental tone existence mark of a current audio frame by utilizing the existing parameters of the long-term post-filter to process the Bluetooth audio in the audio coding process; and in the process of processing the Bluetooth audio by the long-term post-filter, the current pitch delay of the current audio frame is obtained according to the existing processing process, and the current autocorrelation value is obtained. And detecting the audio howling through judging the pitch delay and the autocorrelation value. The pitch lag, autocorrelation value, and the like mentioned in the above process are all processed data in the long-term post-filter processing process, so that the detection of the audio howling and the suppression can be performed without extra operation. Therefore, the complexity of the audio squeal detection process is reduced, the calculation amount is reduced, the power consumption is reduced, and compared with the prior art, the method is more suitable for the field of low-power Bluetooth.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 shows an embodiment of the bluetooth audio howling detection and suppression method of the present application.
In the embodiment shown in fig. 2, the bluetooth audio howling detection and suppression method of the present application includes: in the process S201, in the process of encoding the bluetooth audio by the long-term post-filter, a pitch existence flag of the current audio frame is obtained.
In this embodiment, in the LC3 audio encoder, the long-term post-filter LTPF of the LC3 standard involves processes such as resampling, high-pass filtering, 2-fold down-sampling, pitch detection, and pitch lag estimation when processing the bluetooth audio. According to the method, in the standard processing process of the long-term post-filter to the Bluetooth audio, the output fundamental tone existence mark and the fundamental tone delay are extracted, and then subsequent judgment and processing are carried out. Wherein the Pitch presence flag Pitch _ present indicates whether the current audio frame contains Pitch.
Optionally, judging whether the value of the pitch existence flag is a preset value; if yes, the fact that the current audio frame has fundamental tones is shown, and if not, the fact that the current audio frame does not have the fundamental tones is shown.
In this alternative embodiment, if the Pitch presence flag Pitch _ present is a preset value 1, it indicates that a Pitch exists in the current audio frame, and if the Pitch presence flag Pitch _ present is a preset value 0, it indicates that a Pitch does not exist in the current audio frame.
In the embodiment shown in fig. 2, the bluetooth audio howling detection and suppression method of the present application includes: in the process S202, if the pitch existence flag indicates that the current audio frame has a pitch, the current pitch delay of the current audio frame is determined in the processing process of the long-term post-filter, and the corresponding current autocorrelation value is calculated.
In this embodiment, whether the current audio frame includes a Pitch is determined by using the extracted Pitch presence flag Pitch _ present. And if the Pitch existence flag indicates that the current audio frame has the Pitch, extracting the current Pitch delay Pitch _ delay corresponding to the current audio frame in the processing process of the long-term post-filter. And calculating to obtain the corresponding current autocorrelation value according to the current pitch delay.
Optionally, the calculating to obtain the corresponding current autocorrelation value includes: and in the standard process of calculating the corresponding normalized autocorrelation value according to the current pitch delay, extracting an intermediate calculation value in the standard process as the current autocorrelation value.
In this optional embodiment, the process of calculating the corresponding autocorrelation value according to the pitch lag is similar to the standard process of calculating the normalized autocorrelation value corresponding to the pitch lag in the LC3 encoder, so in an actual process, when calculating the current autocorrelation value, only the corresponding intermediate value needs to be extracted in the calculation process of the standard normalized autocorrelation value, and the current autocorrelation value corresponding to the current audio frame can be obtained.
Specifically, the calculation process of the normalized autocorrelation value of the long-term postfilter of the LC3 audio encoder is shown in the following formula 1:
the calculation process of the current autocorrelation value required by the present application is shown in the following formula 2:
as can be seen from comparing formula 1 with formula 2, formula 1 includes formula 2, so that the autocorrelation value required by the present application can be obtained only in the process of calculating the normalized autocorrelation value specified by the standard, and no additional calculation is required. Wherein, Pitch delay L = Pitch _ delay in the above formula.
Through the above processes, the pitch delay required for the audio howling judgment and the current autocorrelation value corresponding to the current audio frame are both obtained in the standard process of the original long-term post-filter, so that other extra calculation is not needed, and only a program design needs to be properly adjusted. Therefore, compare with the scheme among the prior art, this application operand is low, and is simple, low power consumption, consequently is applicable to the bluetooth low energy field more.
In the embodiment shown in fig. 2, the bluetooth audio howling detection and suppression method of the present application includes: in the process S203, it is determined whether the pitch delays corresponding to the first preset number of previous consecutive audio frames including the current audio frame are the same.
In this optional embodiment, when detecting the bluetooth audio howling, it is first determined whether pitch delays corresponding to a first preset number of encoded audio frames that precede the current audio frame, including the current audio frame, are the same. If the Bluetooth audios are the same, the current Bluetooth audio is indicated to possibly have howling, and subsequent further detection is carried out; if not, the Bluetooth audio is not howled. The first step judgment of the Bluetooth squeal is carried out through the pitch delay, so that the accuracy of the audio squeal detection is improved to a certain extent, and the false detection is avoided.
Specifically, a voice frame of the bluetooth audio is stable for a short time within 10-30 ms, but a pitch period of the voice frame is changed slowly, and if pitch delays of a first preset number of consecutive audio frames are the same, it indicates that there may be howling. The first preset number can be set to be 3-5, wherein the first preset number can be reasonably selected according to the actual detection requirement of the audio howling, and when the first preset number is set to be larger, the requirement on the detection of the audio howling is lower, and the audio howling with a smaller degree is allowed; when the first preset number is set to be smaller, the requirement for detecting the audio howling is higher, and when the audio howling occurs, the audio howling can be found in time to perform a subsequent howling suppression process. The value of the first preset number is not limited in the application.
In the embodiment shown in fig. 2, the bluetooth audio howling detection and suppression method of the present application includes: in the process S204, if the first predetermined number of audio frames are the same, it is determined whether the autocorrelation values corresponding to the first predetermined number of audio frames respectively continuously increase.
In this embodiment, if the pitch lag corresponding to the consecutive first preset number of audio frames is the same, the autocorrelation values of the first preset number of audio frames are determined, and it is determined whether the autocorrelation values continuously increase. If the autocorrelation value continuously increases, the corresponding energy of the audio frame continuously increases, and the characteristic of the typical audio howling indicates that the audio has howling.
Optionally, the determining whether the autocorrelation values respectively corresponding to the first preset number of audio frames continuously increase includes: in the first preset number of audio frames, if the autocorrelation values of the following audio frames are not less than the autocorrelation values of the preceding audio frames, the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase.
In this optional embodiment, when determining whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase, it is determined whether the autocorrelation values of the following audio frames are not less than the autocorrelation values of the preceding audio frames, and if so, it indicates that the autocorrelation values corresponding to the current first preset number of audio frames respectively continuously increase, indicating that there is howling in the audio.
Specifically, when the first preset number is 3, the autocorrelation values of three consecutive audio frames are a, b, and c, respectively, in a front-back order. And when a is more than or equal to b and less than or equal to c, the autocorrelation value is continuously increased, and the audio has howling. It should be noted that, when determining whether the autocorrelation value of the following audio frame is not less than the autocorrelation value of the preceding audio frame, the determination relationship may be set according to different determination accuracy requirements. For example, a < b < c or a < b ≦ c, etc. may be set. In the actual calculation process, because errors of calculation and measurement exist, the actual judgment relationship can be determined according to the requirement of judgment precision, if the judgment precision is high, a is more than b and less than c, and if the judgment precision is low, a is more than or equal to b and less than or equal to c, wherein the specific setting is not specifically limited in the application.
In the embodiment shown in fig. 2, the bluetooth audio howling detection and suppression method of the present application includes: in the process S205, if yes, howling exists in the bluetooth audio, and howling suppression is started.
In this embodiment, when howling is detected in the bluetooth audio, suppression of audio howling is performed from the current audio frame.
Optionally, performing howling suppression, including: in the process of coding the Bluetooth audio by the long-term post-filter, the howling suppression is respectively carried out on a second preset number of post-continuous audio frames including the current audio frame, and the method comprises the following steps: and in the processing process of the long-term post-filter of the second preset number of audio frames, setting the action parameters of the long-term post-filter to be preset values, and writing the preset values into the coded code stream, so that the subsequent decoder does not perform frequency enhancement processing when decoding the second preset number of audio frames.
In this alternative embodiment, when performing howling suppression, a second preset number of consecutive audio frames that are coded next to the current audio frame are subjected to howling suppression processing, starting from the current audio frame. In the long-term post filter, the action parameter Ltpf _ active of the long-term post filter is set to 0 and written into the coded code stream. When the subsequent audio decoder decodes the audio frame, the decoder does not perform related frequency enhancement on the second preset number of audio frames, so that the enhancement of the howling frequency by the decoder is further avoided, and the influence of audio howling is reduced.
Specifically, in the standard LC3 encoding and decoding, if the encoding end detects the pitch, the operation parameter Ltpf _ active of the long-term post filter is set to 1, and the code stream is written, so that the decoding end enhances the relevant pitch frequency component according to the result of the detection of the encoding end. Therefore, if there is howling in the audio, the enhancement of the associated pitch frequency will continue at the decoder side, thus emphasizing the effect of audio howling. Therefore, according to the method and the device, when the audio howling is detected at the encoder end, the action parameter Ltpf _ active of the long-term post-filter is set to 0, so that when the decoder end decodes the audio frame, the enhancement of the related fundamental tone frequency is not performed, the influence of the audio howling is weakened, and the audio howling is suppressed.
It should be noted that the second preset number may be set according to an empirical value of the audio howling suppression. The influence of audio squeal can be effectively reduced, and the audio tone quality is improved.
Specifically, fig. 3 shows an example of the bluetooth audio howling detection and suppression method of the present application.
As shown in fig. 3, in the standard processing procedure of the long-term postfilter, the standard procedures of resampling, high-pass filter filtering, 2-fold down-sampling, pitch detection, and pitch lag estimation are performed in order. Thereby obtaining a pitch presence flag and a pitch delay of the current audio frame. And subsequently judging the existence of the fundamental tone, if the existence flag of the fundamental tone is 1, indicating that the current audio frame contains the fundamental tone, and calculating a corresponding autocorrelation value according to the delay of the fundamental tone. When the pitch delays of the coded continuous N frames including the current audio frame are the same, the fact that howling exists in the current audio is indicated, at the moment, whether the autocorrelation values of the continuous N frames of the audio frame continuously increase or not is judged, and if yes, the fact that howling exists in the audio is indicated. At this time, howling suppression is performed from the current audio frame. The frame number P of the howling suppression attenuation is set, and howling suppression processing is performed on the next P-frame audio frame including the current audio frame. Meanwhile, the action parameter of the long-term post filter is set to be 0 and written into the code stream. Therefore, in the subsequent decoding process, the enhancement of the related fundamental tone frequency can not be carried out, thereby weakening the influence of the audio howling and inhibiting the audio howling.
Optionally, the howling suppression is respectively performed on a second preset number of subsequent consecutive audio frames including the current audio frame, and the method further includes: determining a pitch frequency according to a current pitch delay corresponding to a current audio frame; and setting a notch filter according to the fundamental tone frequency, and filtering a second preset number of audio frames.
In this optional embodiment, the howling suppression process may be performed on a second preset number of subsequent consecutive audio frames including the current audio frame, and a pitch frequency may also be determined according to a current pitch delay corresponding to the current audio frame, and a corresponding notch filter may be further set according to the pitch frequency to filter the second preset number of audio frames and filter clutter in the audio frames, thereby suppressing the howling and improving audio quality.
Specifically, when the pitch frequency is calculated, the pitch lag may be mapped to the pitch frequency based on the relationship of "sampling frequency = pitch lag × pitch frequency", and the input signal may be filtered by arranging a notch filter based on the pitch frequency. Wherein, designing a notch filter according to the known fundamental tone frequency is a mature technology, and the application is not limited. Wherein, the notch filter can select a Finite Impulse Response (FIR) filter or an Infinite Impulse Response (IIR) filter.
Specifically, fig. 4 shows an example of the bluetooth audio howling detection and suppression method of the present application.
As shown in fig. 4, when performing howling suppression, it is first determined whether the number P of audio frames subjected to howling suppression is greater than zero, where P is a second preset number. If yes, subtracting 1 from P, and calculating to obtain corresponding pitch frequency according to the pitch delay of the current audio frame. And sets a corresponding notch filter according to the pitch frequency. And filtering the subsequent P frame audio frame generating the howling through a notch filter, filtering out clutter, reducing the influence of the howling and improving the tone quality.
The method for detecting and inhibiting the Bluetooth audio squeal obtains a fundamental tone existence mark of a current audio frame by utilizing the existing parameters of a long-term post-filter for processing the Bluetooth audio in the audio coding process; and in the process of processing the Bluetooth audio by the long-term post-filter, the current pitch delay of the current audio frame is obtained according to the existing processing process, and the current autocorrelation value is obtained. And detecting the audio howling through judging the pitch delay and the autocorrelation value. The pitch lag, autocorrelation value, and the like mentioned in the above process are intermediate processing data in the long-term post-filter processing process, so that the detection of the audio howling and the suppression can be performed without extra operation. Therefore, the complexity of the audio howling detection process is reduced, the calculation amount is reduced, and the power consumption is reduced. The method can be applied to the Bluetooth low energy, and can also be applied to the classic Bluetooth by matching with the LC3 Bluetooth low energy. The method and the device make full use of the existing information of the audio encoder, assist a small amount of calculation, have low complexity, and save time-frequency conversion which consumes resources greatly compared with the traditional frequency domain howling removing algorithm. Because time-frequency transformation does not exist, a large amount of computation required by the traditional time-frequency transformation is saved, and the method is easy to deploy in embedded equipment. Compare with prior art, be applicable to the bluetooth low energy field more.
Fig. 5 shows an embodiment of the bluetooth audio howling detection and suppression apparatus of the present application.
In the embodiment shown in fig. 5, the bluetooth audio howling detection and suppression apparatus of the present application includes: a howling detection module 501, configured to detect whether a howling exists in the bluetooth audio, where the howling detection module includes: a pitch flag detection module 5011, which obtains a pitch flag of a current audio frame during the coding process of the long-term post-filter on the bluetooth audio; an autocorrelation value obtaining module 5012, configured to determine a current pitch delay of the current audio frame in the processing process of the long-term post-filter when the pitch existence flag indicates that the current audio frame has a pitch, and calculate to obtain a corresponding current autocorrelation value; a judging module 5013, configured to judge whether pitch delays corresponding to a first preset number of previous consecutive audio frames including the current audio frame are the same; if the first preset number of audio frames are the same, judging whether the autocorrelation values corresponding to the first preset number of audio frames respectively continuously increase; if yes, the Bluetooth audio has howling; and a howling suppression module 502, which performs howling suppression on the bluetooth audio in which howling exists.
In this embodiment, the howling detection module is mainly used to detect whether howling exists in the audio. The detection process is completed by the pitch mark detection module, the autocorrelation value acquisition module and the judgment module together. The Pitch flag detection module is used for detecting whether a Pitch exists in a current audio frame, and in a specific detection process, the Pitch existence flag Pitch _ present in a standard long-term post-filter processing process is used for judging. When the Pitch presence flag Pitch _ present is a preset value 1, it indicates that a Pitch is present in the current audio frame, and then subsequent determination is performed.
And in the autocorrelation value acquisition module, extracting the current Pitch delay Pitch _ delay corresponding to the current audio frame in the processing process of the long-term post-filter. And calculating to obtain the corresponding current autocorrelation value according to the current pitch delay. The process of calculating the corresponding autocorrelation value according to the pitch lag is similar to the standard process of calculating the normalized autocorrelation value corresponding to the pitch lag in the LC3 encoder, so in the actual process, the current autocorrelation value only needs to extract the corresponding intermediate value in the calculation process of the standard normalized autocorrelation value, and the current autocorrelation value corresponding to the current audio frame can be obtained.
In the judging module, judging whether the pitch delays corresponding to a first preset number of previous continuous audio frames including the current audio frame are the same or not, if so, indicating that the current Bluetooth audio possibly has squeal, and then carrying out subsequent further detection; if not, the indicated Bluetooth audio does not generate howling. The first step judgment of the Bluetooth squeal is carried out through the pitch delay, so that the accuracy of the audio squeal detection is improved to a certain extent, and the false detection is avoided. If the pitch delays corresponding to the first preset number of previous continuous audio frames are the same, judging whether autocorrelation values corresponding to the first preset number of audio frames are continuously increased or not, if the autocorrelation values are continuously increased, the corresponding energy of the audio frames is continuously increased, and the characteristic of typical audio howling indicates that the audio has howling.
And if the existence of the audio howling is detected, the howling is suppressed through a howling suppression module.
Optionally, the howling suppression module includes a notch filter, and filters the bluetooth audio that generates howling through the notch filter to perform howling suppression.
In this optional embodiment, the howling suppression process may be performed on a second preset number of subsequent consecutive audio frames including the current audio frame, and a pitch frequency may also be determined according to a current pitch delay corresponding to the current audio frame, and a corresponding notch filter may be further set according to the pitch frequency to filter the second preset number of audio frames and filter clutter in the audio frames, thereby suppressing the howling and improving audio quality.
Specifically, when the pitch frequency is calculated, the pitch lag may be mapped to the pitch frequency based on the relationship of "sampling frequency = pitch lag × pitch frequency", and the input signal may be filtered by arranging a notch filter based on the pitch frequency. Wherein, designing a notch filter according to the known fundamental tone frequency is a mature technology, and the application is not limited. Wherein, the notch filter can select a Finite Impulse Response (FIR) filter or an Infinite Impulse Response (IIR) filter.
Specifically, fig. 6 shows an application example of the bluetooth audio howling detection and suppression apparatus at the encoder side according to the present application.
As shown in fig. 6, howling detection and howling suppression are added modules in the present application, and the rest are modules existing in a standard LC3 encoder; determining whether to write related data of a long-term post filter of a code stream or not according to a howling detection result, wherein if howling exists, ltpf _ active =0 is written into the code stream, so that the howling frequency in a decoder is prevented from being further enhanced; if no howling exists, indicating that the fundamental tone is detected in the normal voice, writing the related information into the code stream according to the standard flow, and coding the normal flow. And, the input signal of the subsequent coding, that is, the input of the attach detection and the low-delay modified discrete cosine transform is decided according to whether or not the howling suppression is performed. If howling suppression is executed, the subsequently coded input signal is an audio signal subjected to howling suppression; if howling suppression is not performed, the subsequently encoded input signal is the original audio input signal.
The Bluetooth audio squeal detection and suppression device obtains a fundamental tone existence mark of a current audio frame by utilizing the existing parameters of the long-term post-filter to process the Bluetooth audio in the audio coding process; and in the process of processing the Bluetooth audio by the long-term post-filter, the current pitch delay of the current audio frame is obtained according to the existing processing process, and the current autocorrelation value is obtained. And detecting the audio howling through judging the pitch delay and the autocorrelation value. The pitch lag, autocorrelation value and the like mentioned in the above process are intermediate calculation data in the long-term post-filter processing process, so that the detection of the audio howling can be performed without extra operation, and the suppression can be performed. Therefore, the complexity of the audio squeal detection process is reduced, the calculation amount is reduced, the power consumption is reduced, and compared with the prior art, the method is more suitable for the field of low-power Bluetooth.
Specifically, fig. 7 shows a schematic diagram of an audio signal of howling.
As shown in fig. 7, the simulated howling signal with the sampling rate of 8KHz is shown in the time domain. Wherein, the first half section of the signal is a normal audio signal, and the second half section is a howling signal. As can be seen from the figure, the energy of the signal after the howling signal becomes significantly larger.
Fig. 8 shows a schematic diagram of audio detection corresponding to fig. 7.
As shown in fig. 8, a schematic diagram of a pitch existence flag, a schematic diagram of pitch lag, and autocorrelation values corresponding to the pitch lag in the howling audio in fig. 7 are shown from top to bottom, respectively. As can be seen from fig. 8, in the schematic diagram of the pitch presence flag, when howling occurs in the audio of fig. 7, the corresponding pitch present flag in fig. 7 indicates a 1, and accordingly, the gene delay occurs, specifically 32, wherein in the audio codec of the LC3, the resampling frequency of the long term post-filter is 12.8KHz, if the pitch frequency at which howling occurs is 400Hz, the corresponding pitch delay is 12800/400=32, in the schematic diagram of autocorrelation values, it can be seen that, after the occurrence of audio howling, autocorrelation values show a gradually increasing trend, therefore, in the method and apparatus for detecting and suppressing bluetooth audio howling according to the present application, it is feasible and effective to detect the audio howling through the pitch existence flag, the pitch delay, and the change of the autocorrelation value, and the above processing procedures are only examples, and corresponding processing is performed for different frequency situations.
In a particular embodiment of the present application, a computer-readable storage medium stores computer instructions, wherein the computer instructions are operable to perform the bluetooth audio howling detection and suppression method described in any of the embodiments. Wherein the storage medium may be directly in hardware, in a software module executed by a processor, or in a combination of the two.
A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium.
The Processor may be a Central Processing Unit (CPU), other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), other Programmable logic devices, discrete Gate or transistor logic, discrete hardware components, or any combination thereof. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
In one embodiment of the present application, a computer device includes a processor and a memory, the memory storing computer instructions, wherein: the processor operates the computer instructions to perform the bluetooth audio howling detection and suppression method described in any of the embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above embodiments are merely examples, which are not intended to limit the scope of the present disclosure, and all equivalent structural changes made by using the contents of the specification and the drawings, or any other related technical fields, are also included in the scope of the present disclosure.
Claims (9)
1. A Bluetooth audio howling detection and suppression method, comprising:
in the process of coding the Bluetooth audio by the long-term post filter, acquiring a fundamental tone existence mark of a current audio frame;
if the pitch existence flag indicates that the current audio frame has pitch, determining the current pitch delay of the current audio frame in the processing process of the long-term post-filter, and calculating to obtain a corresponding current autocorrelation value;
judging whether the pitch delays corresponding to the coded continuous first preset number of audio frames including the current audio frame before the current audio frame are the same or not;
if the first preset number of audio frames are the same, judging whether the autocorrelation values of the following audio frames in the first preset number of audio frames are not less than the autocorrelation values of the preceding audio frames;
and if so, the Bluetooth audio has howling, and howling suppression is started.
2. The method of claim 1, wherein the step of detecting and suppressing the bluetooth audio howling comprises, if the pitch presence flag indicates that a pitch is present in the current audio frame:
judging whether the value of the pitch existence flag is a preset value or not;
if yes, the current audio frame has a fundamental tone, and if not, the current audio frame has no fundamental tone.
3. The method according to claim 1, wherein the calculating for obtaining the current autocorrelation value corresponding thereto comprises:
and in a standard process of calculating the corresponding normalized autocorrelation value according to the current pitch delay, extracting an intermediate calculation value of the standard process to be used as the current autocorrelation value.
4. The bluetooth audio howling detection and suppression method according to claim 1, wherein said performing howling suppression comprises:
during the encoding process of the long-term post-filter on the bluetooth audio, performing howling suppression on a second preset number of consecutive audio frames, which include the current audio frame and are to be encoded after the current audio frame, respectively, including:
and in the processing process of the long-term post-filter of the second preset number of audio frames, setting the action parameters of the long-term post-filter to be preset values, and writing the preset values into the coded code stream, so that a subsequent decoder does not perform frequency enhancement processing when decoding the second preset number of audio frames.
5. The bluetooth audio howling detection and suppression method according to claim 4, wherein said howling suppression is respectively performed on a second preset number of subsequent consecutive audio frames including the current audio frame, further comprising:
determining a pitch frequency according to the current pitch delay corresponding to the current audio frame;
and setting a notch filter according to the fundamental tone frequency, and filtering the second preset number of audio frames.
6. A bluetooth audio howling detection and suppression apparatus, comprising:
the howling detection module detects whether howling exists in the Bluetooth audio, and comprises:
a pitch mark detection module, which acquires a pitch mark of the current audio frame in the process of coding the Bluetooth audio by the long-term post filter;
an autocorrelation value obtaining module, configured to determine a current pitch delay of the current audio frame in a processing process of the long-term post-filter when the pitch existence flag indicates that the current audio frame has a pitch, and calculate to obtain a corresponding current autocorrelation value;
the judging module is used for judging whether the pitch delays corresponding to the coded continuous first preset number of audio frames before the current audio frame, including the current audio frame, are the same or not;
if the first preset number of audio frames are the same, judging whether the autocorrelation values of the following audio frames in the first preset number of audio frames are not less than the autocorrelation values of the preceding audio frames;
if yes, the Bluetooth audio has howling;
and the howling suppression module is used for carrying out howling suppression on the Bluetooth audio with howling.
7. The bluetooth audio howling detection and suppression device according to claim 6, wherein said howling suppression module comprises:
and the notch filter is used for filtering the Bluetooth audio which generates the howling through the notch filter so as to carry out howling suppression.
8. A computer readable storage medium storing computer instructions, wherein the computer instructions are operable to perform the bluetooth audio howling detection and suppression method of any one of claims 1-5.
9. A computer device comprising a processor and a memory, the memory storing computer instructions, wherein: the processor operates computer instructions to perform the bluetooth audio howling detection and suppression method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111456960.5A CN113870885B (en) | 2021-12-02 | 2021-12-02 | Bluetooth audio squeal detection and suppression method, device, medium, and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111456960.5A CN113870885B (en) | 2021-12-02 | 2021-12-02 | Bluetooth audio squeal detection and suppression method, device, medium, and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113870885A CN113870885A (en) | 2021-12-31 |
| CN113870885B true CN113870885B (en) | 2022-02-22 |
Family
ID=78985628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111456960.5A Active CN113870885B (en) | 2021-12-02 | 2021-12-02 | Bluetooth audio squeal detection and suppression method, device, medium, and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113870885B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114067814B (en) * | 2022-01-18 | 2022-04-12 | 北京百瑞互联技术有限公司 | Howling detection and suppression method and device based on Bluetooth audio receiver |
| CN114420148B (en) * | 2022-03-30 | 2022-06-14 | 北京百瑞互联技术有限公司 | Howling detection and suppression method, device, medium and equipment |
| CN114501282B (en) * | 2022-04-06 | 2022-06-24 | 武汉左点科技有限公司 | External squeal suppression method and device for hearing aid |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101228573A (en) * | 2005-07-27 | 2008-07-23 | 摩托罗拉公司 | Method and apparatus for coding an information signal using pitch delay contour adjustment |
| CN109640223A (en) * | 2019-02-22 | 2019-04-16 | 昆腾微电子股份有限公司 | A kind of chauvent's criterion method, apparatus, sound equipment and public address system |
| CN110838301A (en) * | 2019-11-20 | 2020-02-25 | 北京雷石天地电子技术有限公司 | Method, device terminal and non-transitory computer readable storage medium for suppressing howling |
| CN210431880U (en) * | 2019-08-16 | 2020-04-28 | 福建太尔集团股份有限公司 | Bone conduction hearing aid with squeaking prevention function |
| CN111477246A (en) * | 2019-01-24 | 2020-07-31 | 腾讯科技（深圳）有限公司 | Voice processing method and device and intelligent terminal |
| CN112289328A (en) * | 2020-10-28 | 2021-01-29 | 北京百瑞互联技术有限公司 | Method and system for determining audio coding rate |
| CN113724725A (en) * | 2021-11-04 | 2021-11-30 | 北京百瑞互联技术有限公司 | Bluetooth audio squeal detection suppression method, device, medium and Bluetooth device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB0204057D0 (en) * | 2002-02-21 | 2002-04-10 | Tecteon Plc | Echo detector having correlator with preprocessing |
| JP4649888B2 (en) * | 2004-06-24 | 2011-03-16 | ヤマハ株式会社 | Voice effect imparting device and voice effect imparting program |
| NO332437B1 (en) * | 2010-01-18 | 2012-09-17 | Cisco Systems Int Sarl | Apparatus and method for suppressing an acoustic echo |
| US9628923B2 (en) * | 2013-12-27 | 2017-04-18 | Gn Hearing A/S | Feedback suppression |
| US10708689B2 (en) * | 2018-05-15 | 2020-07-07 | LogMeln, Inc. | Reducing acoustic feedback over variable-delay pathway |
-
2021
- 2021-12-02 CN CN202111456960.5A patent/CN113870885B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101228573A (en) * | 2005-07-27 | 2008-07-23 | 摩托罗拉公司 | Method and apparatus for coding an information signal using pitch delay contour adjustment |
| CN111477246A (en) * | 2019-01-24 | 2020-07-31 | 腾讯科技（深圳）有限公司 | Voice processing method and device and intelligent terminal |
| CN109640223A (en) * | 2019-02-22 | 2019-04-16 | 昆腾微电子股份有限公司 | A kind of chauvent's criterion method, apparatus, sound equipment and public address system |
| CN210431880U (en) * | 2019-08-16 | 2020-04-28 | 福建太尔集团股份有限公司 | Bone conduction hearing aid with squeaking prevention function |
| CN110838301A (en) * | 2019-11-20 | 2020-02-25 | 北京雷石天地电子技术有限公司 | Method, device terminal and non-transitory computer readable storage medium for suppressing howling |
| CN112289328A (en) * | 2020-10-28 | 2021-01-29 | 北京百瑞互联技术有限公司 | Method and system for determining audio coding rate |
| CN113724725A (en) * | 2021-11-04 | 2021-11-30 | 北京百瑞互联技术有限公司 | Bluetooth audio squeal detection suppression method, device, medium and Bluetooth device |
Non-Patent Citations (2)
| Title |
|---|
| "Acoustic echo control for high quality audio teleconferencing";Yasukawa H;《ICASSP》;20021231;全文 * |
| "数学形态学在语音识别中的应用研究";王霞;《中国博士学位论文全文数据库信息科技辑》;20090815;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113870885A (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113870885B (en) | Bluetooth audio squeal detection and suppression method, device, medium, and apparatus | |
| CN113724725B (en) | Bluetooth audio squeal detection suppression method, device, medium and Bluetooth device | |
| JP6412132B2 (en) | Voice activity detection method and apparatus | |
| CN111383646B (en) | Voice signal transformation method, device, equipment and storage medium | |
| CN111681663B (en) | Method, system, storage medium and device for reducing audio coding computation amount | |
| CN112309414B (en) | Active noise reduction method based on audio encoding and decoding, earphone and electronic equipment | |
| CN114550732B (en) | Coding and decoding method and related device for high-frequency audio signal | |
| CN111754983A (en) | Voice denoising method and device, electronic equipment and storage medium | |
| CN110610696B (en) | MFCC feature extraction method and device based on mixed signal domain | |
| CN112967738A (en) | Human voice detection method and device, electronic equipment and computer readable storage medium | |
| JP4551817B2 (en) | Noise level estimation method and apparatus | |
| CN111477246B (en) | Voice processing method and device and intelligent terminal | |
| CN114067814B (en) | Howling detection and suppression method and device based on Bluetooth audio receiver | |
| CN105355206B (en) | Voiceprint feature extraction method and electronic equipment | |
| CN113223561B (en) | Voice activity detection method, electronic equipment and device | |
| JP2010102129A (en) | Fundamental frequency extracting method, fundamental frequency extracting device, and program | |
| CN116312616A (en) | Processing recovery method and control system for noisy speech signals | |
| CN115273880A (en) | Voice noise reduction method, model training method, device, equipment, medium and product | |
| CN106340310A (en) | Speech detection method and device | |
| JP2015161718A (en) | speech detection device, speech detection method and speech detection program | |
| CN104599682A (en) | Method for extracting pitch period of telephone wire quality voice | |
| CN114420148B (en) | Howling detection and suppression method, device, medium and equipment | |
| CN113823277A (en) | Keyword recognition method, system, medium, and apparatus based on deep learning | |
| CN113905310B (en) | Bluetooth audio squeal detection and suppression method, device and medium | |
| CN112489692A (en) | Voice endpoint detection method and device |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: A1009, floor 9, block a, No. 9, Shangdi Third Street, Haidian District, Beijing 100085 Patentee after: Beijing Bairui Internet Technology Co.,Ltd. Address before: A1009, floor 9, block a, No. 9, Shangdi Third Street, Haidian District, Beijing 100085 Patentee before: BARROT WIRELESS Co.,Ltd. |