CN109102819A - One kind is uttered long and high-pitched sounds detection method and device - Google Patents
One kind is uttered long and high-pitched sounds detection method and device Download PDFInfo
- Publication number
- CN109102819A CN109102819A CN201710468828.3A CN201710468828A CN109102819A CN 109102819 A CN109102819 A CN 109102819A CN 201710468828 A CN201710468828 A CN 201710468828A CN 109102819 A CN109102819 A CN 109102819A
- Authority
- CN
- China
- Prior art keywords
- howling
- signal
- threshold value
- voice input
- energy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- 230000001131 transforming effect Effects 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 33
- 230000006870 function Effects 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000009432 framing Methods 0.000 claims 2
- 230000001629 suppression Effects 0.000 abstract description 9
- 230000000875 corresponding effect Effects 0.000 description 18
- 230000002787 reinforcement Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 10
- 238000004590 computer program Methods 0.000 description 5
- 230000002596 correlated effect Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques 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 TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques 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
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L21/0232—Processing in the frequency domain
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
本发明实施例提供一种啸叫检测方法及装置，用于解决现有技术存在检测啸叫不精准的技术问题。所述啸叫检测方法包括：将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；根据扬声器的语音输入信号的能量确定啸叫门限值；判断所述第一比值是否大于所述啸叫门限值；若为是，则确定所述第一频点为啸叫频点，并对所述第一频点进行啸叫抑制。
Embodiments of the present invention provide a howling detection method and device, which are used to solve the technical problem of inaccurate detection of howling in the prior art. The howling detection method includes: transforming the voice signal collected by the microphone from the time domain to the frequency domain to obtain a frequency domain signal corresponding to the voice signal; calculating a first signal energy of the frequency domain signal at a first frequency point The first ratio with the total energy of the frequency domain signal, the first frequency point is any frequency point in the frequency domain where the frequency domain signal is located; determine the howling threshold value according to the energy of the voice input signal of the speaker; determine Whether the first ratio is greater than the howling threshold; if yes, determine that the first frequency point is a howling frequency point, and perform howling suppression on the first frequency point.
Description
技术领域technical field
本发明涉及信号处理领域，特别涉及一种啸叫检测方法及装置。The invention relates to the field of signal processing, in particular to a howling detection method and device.
背景技术Background technique
在扩声系统中，扬声器播放的语音信号经过障碍物反射后会返回到麦克风被麦克风再次拾音，然后麦克风将反射回来的语音信号送往扬声器播放，这样周而复始的叠加使得部分语音信号在扩声系统中发生自激振荡，发出刺耳的尖叫声，即啸叫。In the sound reinforcement system, the voice signal played by the speaker will return to the microphone to be picked up again by the microphone after being reflected by obstacles, and then the microphone will send the reflected voice signal to the speaker for playback, so that the repeated superposition makes part of the voice signal in the sound reinforcement Self-excited oscillation occurs in the system, and a piercing scream is issued, that is, howling.
为了抑制啸叫的发生，需要对信号进行啸叫检测，现有技术检测啸叫的通常做法为：对麦克风采集到的语音信号进行频域处理，然后根据该语音信号的能量的特征值设定一个阈值，并将该语音信号各频点处信号的能量的特征值与该阈值进行比较，以此判断该语音信号是否存在啸叫频点。In order to suppress the occurrence of howling, it is necessary to perform howling detection on the signal. The usual way to detect howling in the prior art is to process the voice signal collected by the microphone in the frequency domain, and then set A threshold, and compare the characteristic value of the energy of the signal at each frequency point of the voice signal with the threshold, so as to judge whether there is a howling frequency point in the voice signal.
但是，现有技术上述检测啸叫频点的技术方案的精准较差。However, the above-mentioned technical solutions for detecting howling frequency points in the prior art are relatively poor in precision.
发明内容Contents of the invention
本发明提供一种信息处理方法及电子设备，用于解决现有技术存在检测啸叫频点不精准的技术问题。The invention provides an information processing method and electronic equipment, which are used to solve the technical problem of inaccurate detection of howling frequency points in the prior art.
本发明实施例第一方面一种啸叫检测方法，包括：In the first aspect of the embodiments of the present invention, a howling detection method includes:
将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；Transforming the voice signal collected by the microphone from the time domain to the frequency domain to obtain a frequency domain signal corresponding to the voice signal;
计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；calculating a first ratio between the first signal energy of the frequency domain signal at a first frequency point and the total energy of the frequency domain signal, where the first frequency point is any frequency within the frequency domain where the frequency domain signal is located point;
根据扬声器的语音输入信号的能量确定啸叫门限值；determining the howling threshold value according to the energy of the voice input signal of the loudspeaker;
判断所述第一比值是否大于所述啸叫门限值；judging whether the first ratio is greater than the howling threshold;
若为是，则确定所述第一频点为啸叫频点。If yes, it is determined that the first frequency point is a howling frequency point.
在上述方案中，根据麦克风采集的语音信号和扬声器的语音输入信号确定啸叫频点，能够使得检测出的啸叫频点更加准确，进而使得对啸叫的抑制效果更好，提升了扩声系统的性能。In the above solution, determining the howling frequency point according to the voice signal collected by the microphone and the voice input signal of the speaker can make the detected howling frequency point more accurate, thereby making the howling suppression effect better and improving the sound reinforcement. system performance.
可选的，所述根据扬声器的语音输入信号的能量确定啸叫门限值，包括：判断所述扬声器的语音输入信号的能量是否大于0；若为是，则确定所述扬声器有语音输入，并确定所述啸叫门限值为第一啸叫门限值，所述第一啸叫门限值大于0且小于1；若为否，则确定所述扬声器无语音输入，并确定所述啸叫门限值为第二啸叫门限值，所述第二啸叫门限值大于所述第一啸叫门限值且小于1。通过本方式，在啸叫发生概率较大的情况下减小啸叫门限值以及在啸叫发生概率较小的情况下增大啸叫门限值，实现了对啸叫门限值的灵活调整，进一步提高了啸叫检测的准确性。Optionally, the determining the howling threshold value according to the energy of the voice input signal of the speaker includes: judging whether the energy of the voice input signal of the speaker is greater than 0; if yes, determining that the speaker has voice input, and Determining that the howling threshold is a first howling threshold, and the first howling threshold is greater than 0 and less than 1; if not, then determining that the speaker has no voice input, and determining the howling threshold is a second howling threshold, and the second howling threshold is greater than the first howling threshold and less than 1. Through this method, the howling threshold value is reduced when the howling occurrence probability is high and the howling threshold value is increased when the howling occurrence probability is small, so that the flexible adjustment of the howling threshold value is realized, and further Improved the accuracy of howling detection.
可选的，所述根据扬声器的语音输入信号的能量确定啸叫门限值，包括：在所述扬声器的语音输入信号的平均能量属于第一预定范围时，确定所述啸叫门限值为第三啸叫门限值，所述扬声器的语音输入信号的平均能量等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，所述第三啸叫门限值大于0且小于1；在所述扬声器的语音输入信号的平均能量属于第二预定范围时，确定所述啸叫门限值为第四啸叫门限值，所述第二预定范围内的最小值大于等于0，所述第二预定范围内的最大值小于所述第一预定范围内的最小值，所述第四啸叫门限值大于所述第三啸叫门限值且小于1。通过本方式，在啸叫发生概率较大的情况下减小啸叫门限值以及在啸叫发生概率较小的情况下增大啸叫门限值，实现了对啸叫门限值的灵活调整，进一步提高了啸叫检测的准确性。Optionally, the determining the howling threshold value according to the energy of the voice input signal of the speaker includes: when the average energy of the voice input signal of the speaker belongs to a first predetermined range, determining the howling threshold value as a third Howling threshold value, the average energy of the voice input signal of the speaker is equal to the energy of the voice input signal of the speaker divided by the duration of the voice input signal of the speaker, the third howling threshold value is greater than 0 and less than 1 ; when the average energy of the voice input signal of the speaker belongs to a second predetermined range, determine that the howling threshold is a fourth howling threshold, the minimum value in the second predetermined range is greater than or equal to 0, and the The maximum value within the second predetermined range is smaller than the minimum value within the first predetermined range, and the fourth howling threshold is greater than the third howling threshold and less than 1. Through this method, the howling threshold value is reduced when the howling occurrence probability is high and the howling threshold value is increased when the howling occurrence probability is small, so that the flexible adjustment of the howling threshold value is realized, and further Improved the accuracy of howling detection.
可选的，所述根据扬声器的语音输入信号的能量确定啸叫门限值，包括：基于公式：y＝f(x)确定所述啸叫门限值；其中，y为所述啸叫门限值，x为所述扬声器的语音输入信号的平均能量，x等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，f(x)表示以x为变量的函数，f(x)大于0且小于1，所述啸叫门限值y与单位时间内所述扬声器的语音输入信号的能量x呈负相关关系。通过本方式，在啸叫发生概率较大的情况下减小啸叫门限值以及在啸叫发生概率较小的情况下增大啸叫门限值，实现了对啸叫门限值的灵活调整，进一步提高了啸叫检测的准确性。Optionally, the determining the howling threshold value according to the energy of the voice input signal of the speaker includes: determining the howling threshold value based on the formula: y=f(x); wherein, y is the howling threshold value, x is the average energy of the speech input signal of the loudspeaker, x equals the energy of the speech input signal of the loudspeaker divided by the duration of the speech input signal of the loudspeaker, f(x) represents a function with x as a variable, f( x) is greater than 0 and less than 1, and the howling threshold y is negatively correlated with the energy x of the voice input signal of the loudspeaker per unit time. Through this method, the howling threshold value is reduced when the howling occurrence probability is high and the howling threshold value is increased when the howling occurrence probability is small, so that the flexible adjustment of the howling threshold value is realized, and further Improved the accuracy of howling detection.
可选的，所述扬声器的语音输入信号为所述扬声器在超过第一阈值的时长范围内接收的语音输入信号。通过本方式，能够使得啸叫门限值更加准确，进而使得检测出的啸叫频点更加准确。Optionally, the voice input signal of the speaker is a voice input signal received by the speaker within a time range exceeding a first threshold. Through this method, the howling threshold value can be made more accurate, thereby making the detected howling frequency points more accurate.
可选的，所述将麦克风采集的语音信号从时域变换到频域，包括：将所述麦克风采集的语音信号进行分帧处理，得到M帧时域信号，M为大于等于1的正整数；对所述M帧时域信号中的任意一帧时域信号进行快速傅里叶变换FTT处理，得到所述频域信号。通过本方式将语音信号进行分帧处理，并针对每一帧语音信号单独进行啸叫检测，可以提升啸叫检测效率，使得检测出的啸叫频点更加准确。Optionally, said transforming the voice signal collected by the microphone from the time domain to the frequency domain includes: performing frame division processing on the voice signal collected by the microphone to obtain M frames of time domain signals, where M is a positive integer greater than or equal to 1 ; performing fast Fourier transform (FTT) processing on any frame of time domain signals in the M frames of time domain signals to obtain the frequency domain signal. In this manner, the voice signal is processed in frames, and the howling detection is performed separately for each frame of the voice signal, which can improve the howling detection efficiency and make the detected howling frequency points more accurate.
本发明实施例第二方面提供一种啸叫检测装置，包括：转换单元，用于将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；计算单元，用于计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；第一确定单元，用于根据扬声器的语音输入信号的能量确定啸叫门限值；判断单元，用于判断所述第一比值是否大于所述啸叫门限值；第二确定单元，用于在所述第一比值大于所述啸叫门限值时，确定所述第一频点为啸叫频点。The second aspect of the embodiment of the present invention provides a howling detection device, including: a conversion unit, configured to convert a voice signal collected by a microphone from the time domain to the frequency domain, to obtain a frequency domain signal corresponding to the voice signal; a calculation unit, It is used to calculate a first ratio between the first signal energy of the frequency domain signal at a first frequency point and the total energy of the frequency domain signal, and the first frequency point is any frequency domain within the frequency domain where the frequency domain signal is located. A frequency point; the first determining unit is used to determine the howling threshold value according to the energy of the voice input signal of the speaker; the judging unit is used to judge whether the first ratio is greater than the howling threshold value; the second determining unit, It is used for determining that the first frequency point is a howling frequency point when the first ratio is greater than the howling threshold value.
可选的，所述第一确定单元用于判断所述扬声器的语音输入信号的能量是否大于0；在所述扬声器的语音输入信号的能量大于0时确定所述扬声器有语音输入，并确定所述啸叫门限值为第一啸叫门限值，所述第一啸叫门限值大于0且小于1；在所述扬声器的语音输入信号的能量不大于0时确定所述扬声器无语音输入，并确定所述啸叫门限值为第二啸叫门限值，所述第二啸叫门限值大于所述第一啸叫门限值且小于1。Optionally, the first determination unit is configured to determine whether the energy of the voice input signal of the speaker is greater than 0; when the energy of the voice input signal of the speaker is greater than 0, determine that the speaker has voice input, and determine that the The howling threshold value is a first howling threshold value, and the first howling threshold value is greater than 0 and less than 1; when the energy of the voice input signal of the speaker is not greater than 0, it is determined that the speaker has no voice input, and Determine that the howling threshold is a second howling threshold, where the second howling threshold is greater than the first howling threshold and less than 1.
可选的，所述第一确定单元用于在所述扬声器的语音输入信号的平均能量属于第一预定范围时，确定所述啸叫门限值为第三啸叫门限值，所述扬声器的语音输入信号的平均能量等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，所述第三啸叫门限值大于0且小于1；在所述扬声器的语音输入信号的平均能量属于第二预定范围时，确定所述啸叫门限值为第四啸叫门限值，所述第二预定范围内的最小值大于等于0，所述第二预定范围内的最大值小于所述第一预定范围内的最小值，所述第四啸叫门限值大于所述第三啸叫门限值且小于1。Optionally, the first determination unit is configured to determine that the howling threshold is a third howling threshold when the average energy of the voice input signal of the speaker falls within a first predetermined range, and the voice of the speaker is The average energy of the input signal is equal to the energy of the voice input signal of the loudspeaker divided by the duration of the voice input signal of the loudspeaker, and the third howling threshold is greater than 0 and less than 1; When the average energy belongs to the second predetermined range, it is determined that the howling threshold is a fourth howling threshold, the minimum value in the second predetermined range is greater than or equal to 0, and the maximum value in the second predetermined range is less than the set value. The minimum value within the first predetermined range, the fourth howling threshold is greater than the third howling threshold and less than 1.
可选的，所述第一确定单元用于基于公式：y＝f(x)确定所述啸叫门限值；其中，y为所述啸叫门限值，x为所述扬声器的语音输入信号的平均能量，x等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，f(x)表示以x为变量的函数，f(x)大于0且小于1，所述啸叫门限值y与单位时间内所述扬声器的语音输入信号的能量x呈负相关关系。Optionally, the first determination unit is configured to determine the howling threshold based on the formula: y=f(x); wherein, y is the howling threshold, and x is the voice input signal of the speaker Average energy, x is equal to the energy of the voice input signal of the speaker divided by the duration of the voice input signal of the speaker, f(x) represents a function with x as a variable, f(x) is greater than 0 and less than 1, the The howling threshold value y is negatively correlated with the energy x of the voice input signal of the loudspeaker per unit time.
可选的，所述扬声器的语音输入信号为所述扬声器在超过第一阈值的时长范围内接收的语音输入信号。Optionally, the voice input signal of the speaker is a voice input signal received by the speaker within a time range exceeding a first threshold.
可选的，所述转换单元用于将所述麦克风采集的语音信号进行分帧处理，得到M帧时域信号，M为大于等于1的正整数；对所述M帧时域信号中的任意一帧时域信号进行快速傅里叶变换FTT处理，得到所述频域信号。Optionally, the conversion unit is configured to process the voice signal collected by the microphone into frames to obtain M frames of time domain signals, where M is a positive integer greater than or equal to 1; for any of the M frames of time domain signals A frame of time-domain signals is subjected to fast Fourier transform (FTT) processing to obtain the frequency-domain signals.
本发明实施例第三方面提供一种啸叫检测系统，所述系统包括拾音电路、信号处理电路以及信号判断电路；所述拾音电路用于获取麦克风采集的语音信号和扬声器的语音输入信号；所述信号处理电路与所述拾音电路相连，用于将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；所述信号判断电路与所述信号处理电路相连，用于计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；根据扬声器的语音输入信号的能量确定啸叫门限值；判断所述第一比值是否大于所述啸叫门限值；在所述第一比值大于所述啸叫门限值时，确定所述第一频点为啸叫频点。The third aspect of the embodiment of the present invention provides a howling detection system, the system includes a sound pickup circuit, a signal processing circuit and a signal judgment circuit; the sound pickup circuit is used to obtain the voice signal collected by the microphone and the voice input signal of the speaker The signal processing circuit is connected with the sound pickup circuit, and is used to convert the voice signal collected by the microphone from the time domain to the frequency domain, so as to obtain the frequency domain signal corresponding to the voice signal; the signal judging circuit and the signal The processing circuit is connected to calculate a first ratio between the first signal energy of the frequency domain signal at a first frequency point and the total energy of the frequency domain signal, and the first frequency point is where the frequency domain signal is located Any frequency point in the frequency domain; determine the howling threshold value according to the energy of the voice input signal of the loudspeaker; judge whether the first ratio is greater than the howling threshold value; when the first ratio is greater than the howling threshold value , it is determined that the first frequency point is a howling frequency point.
本发明实施例第四方面提供一种啸叫检测设备，包括：至少一个处理器；以及与所述至少一个处理器连接的存储器；其中，所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行本发明实施例第一方面提供的所述啸叫检测方法。The fourth aspect of the embodiment of the present invention provides a howling detection device, including: at least one processor; and a memory connected to the at least one processor; wherein, the memory stores information that can be executed by the at least one processor. instructions, the instructions are executed by the at least one processor, so that the at least one processor can execute the howling detection method provided in the first aspect of the embodiments of the present invention.
附图说明Description of drawings
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简要介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without any creative effort.
图1为本发明实施例提供的啸叫检测方法的流程示意图；FIG. 1 is a schematic flowchart of a howling detection method provided by an embodiment of the present invention;
图2为本发明实施例中扬声器的语音输入信号的平均能量与啸叫门限值的的示意图；FIG. 2 is a schematic diagram of the average energy and howling threshold of the voice input signal of the speaker in an embodiment of the present invention;
图3为本发明实施例中扬声器的语音输入信号的平均能量与啸叫门限值的的示意图；3 is a schematic diagram of the average energy and howling threshold of the voice input signal of the speaker in an embodiment of the present invention;
图4为本发明实施例中扬声器的语音输入信号的平均能量与啸叫门限值的的示意图；4 is a schematic diagram of the average energy and howling threshold of the voice input signal of the speaker in an embodiment of the present invention;
图5为本发明实施例提供的啸叫检测装置的结构示意图；FIG. 5 is a schematic structural diagram of a howling detection device provided by an embodiment of the present invention;
图6为本发明实施例提供的啸叫检测系统的结构示意图；FIG. 6 is a schematic structural diagram of a howling detection system provided by an embodiment of the present invention;
图7为本发明实施例提供的啸叫检测设备的结构示意图。Fig. 7 is a schematic structural diagram of a howling detection device provided by an embodiment of the present invention.
具体实施方式Detailed ways
下面通过附图以及具体实施例对本发明技术方案做详细的说明，应当理解本发明实施例以及实施例中的具体特征是对本发明技术方案的详细的说明，而不是对本发明技术方案的限定，在不冲突的情况下，本发明实施例以及实施例中的技术特征可以相互组合。The technical solutions of the present invention will be described in detail below through the accompanying drawings and specific examples. It should be understood that the embodiments of the present invention and the specific features in the examples are detailed descriptions of the technical solutions of the present invention, rather than limitations to the technical solutions of the present invention. In the case of no conflict, the embodiments of the present invention and the technical features in the embodiments may be combined with each other.
为了抑制啸叫的发生，现有技术检测啸叫的通常做法是对麦克风采集的语音信号本身特征进行分析，人为地设定一个阈值，然后将该语音信号各频点处信号的能量的特征值与该阈值进行比较，以此判断该语音信号是否存在啸叫频点，并针对啸叫频点采取啸叫抑制措施。但是，在扬声器处于不同输出状态时，麦克风采集到的从扬声器处出来的语音信号不同，发生啸叫的可能性也不同，如果在扬声器处于不同输出状态下用同一阈值去检测啸叫，很难检测出准确的啸叫频点。In order to suppress the occurrence of howling, the usual method of detecting howling in the prior art is to analyze the characteristics of the voice signal itself collected by the microphone, artificially set a threshold, and then use the characteristic value of the energy of the signal at each frequency point of the voice signal Compared with the threshold value, it is judged whether there is a howling frequency point in the voice signal, and howling suppression measures are taken for the howling frequency point. However, when the speaker is in different output states, the voice signals collected by the microphone from the speaker are different, and the possibility of howling may occur is also different. If the speaker is in different output states with the same threshold to detect howling, it is difficult Detect the exact howling frequency point.
实施例一Embodiment one
本发明实施例一提供一种啸叫检测方法，用于解决现有技术存在检测啸叫不精准的技术问题。参照图1，所述方法包括：Embodiment 1 of the present invention provides a howling detection method, which is used to solve the technical problem of inaccurate detection of howling in the prior art. Referring to Figure 1, the method includes:
步骤110：将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号。Step 110: Transform the voice signal collected by the microphone from the time domain to the frequency domain to obtain a frequency domain signal corresponding to the voice signal.
具体的，获取麦克风在一段时长内的所采集的语音信号，并对获取到的语音信号进行分帧处理，得到M帧时域信号；然后再对M帧时域信号中的每一帧时域信号分别进行频域分带处理，得到M个频域信号；最后将所述M个频域信号中的任意一个频域信号作为所述语音信号对应的频域信号。其中，每个频域信号所在的频带包括至少两个频带区间，一个频带区间表征一个频点。Specifically, the voice signal collected by the microphone within a period of time is obtained, and the acquired voice signal is processed in frames to obtain M frames of time domain signals; and then each frame of time domain signals in the M frames of time domain signals is The signals are respectively subjected to frequency-domain band-division processing to obtain M frequency-domain signals; finally, any one of the M frequency-domain signals is used as a frequency-domain signal corresponding to the speech signal. Wherein, the frequency band where each frequency domain signal is located includes at least two frequency band intervals, and one frequency band interval represents one frequency point.
其中，对M帧时域信号分别进行频域分带处理可以有多种实现方式，包括但不限于以下两种：Wherein, there are many ways to implement the frequency-domain band-splitting processing on the M-frame time-domain signals, including but not limited to the following two:
第1种，可以先对M帧时域信号的每一帧时域信号进行加窗处理，然后将加窗后的时域信号分别进行快速傅里叶变换(fast Fourier transform，FTT)处理，根据一帧时域信号得到一个频域信号，一共可以得到M个频域信号，然后再对M个频域信号中的每一频域信号进行频带划分处理，即将频域信号所在的频带划分为至少两个频带区间，一个频带区间表征一个频点，其中不同频域信号所在的频带的频带区间划分方式可以相同，也可以不同，本发明实施例不做具体限制。In the first type, windowing processing can be performed on each frame of time domain signals of M frames of time domain signals first, and then fast Fourier transform (FTT) processing is performed on the time domain signals after windowing, according to A frame of time-domain signal obtains a frequency-domain signal, a total of M frequency-domain signals can be obtained, and then each frequency-domain signal in the M frequency-domain signals is divided into frequency bands, that is, the frequency band where the frequency-domain signal is located is divided into at least There are two frequency band intervals, one frequency band interval represents one frequency point, and the frequency band interval division methods of the frequency bands where different frequency domain signals are located may be the same or different, which is not specifically limited in the embodiment of the present invention.
例如，获取麦克风在360ms内采集的语音信号，然后按照20ms为一帧将该段语音信号进行划分，共得到18帧时域信号；然后，对所述18帧时域信号中的每一帧时域信号分别进行FTT处理，可以得到18个对应的频域信号；最后，对18个频域信号中的每个频域信号进行频带划分处理，将每个频域信号所在的频带划分为128个频带区间，即128个频点。当然了，上述只是一个特例，在具体实施过程中，上述语音信号还可以是麦克风在任意时间长度内采集的语音信号，例如300ms、400ms、500ms等，本发明实施例不做具体限制；上述对获取到的语音信号进行分帧处理还可以有其它实现方式，例如按照10ms为一帧、12ms为一帧、30ms为一帧等，将该段语音信号进行划分，本发明实施例对此不做具体限制。For example, obtain the speech signal collected by the microphone within 360ms, then divide the speech signal according to 20ms as a frame, and obtain 18 frames of time domain signals; then, each frame of the 18 frames of time domain signals Domain signals are processed by FTT respectively, and 18 corresponding frequency domain signals can be obtained; finally, each frequency domain signal in the 18 frequency domain signals is divided into frequency bands, and the frequency band of each frequency domain signal is divided into 128 Frequency band interval, that is, 128 frequency points. Of course, the above is only a special case. In the specific implementation process, the above-mentioned voice signal can also be a voice signal collected by the microphone within any length of time, such as 300ms, 400ms, 500ms, etc., and the embodiment of the present invention does not make specific limitations; the above-mentioned There may also be other implementations for the frame division processing of the acquired voice signal. For example, according to 10ms as a frame, 12ms as a frame, and 30ms as a frame, etc., the segment of the voice signal is divided, and the embodiment of the present invention does not do this Specific restrictions.
第2种，采用多个不同频段的滤波器组合对M帧时域信号的每一帧时域信号进行频域分带处理，得到所述M个频域信号，其中每个频域信号所在的频带包括至少两个频带区间，一个频带区间表征一个频点。In the second type, multiple filter combinations of different frequency bands are used to perform frequency-domain band-division processing on each frame of time-domain signals of M frames of time-domain signals to obtain the M frequency-domain signals, where each frequency-domain signal is located The frequency band includes at least two frequency band intervals, and one frequency band interval represents one frequency point.
可选的，本发明在具体实施过程中也可以不对麦克风采集的语音信号进行分帧处理，而是将该语音信号作为一帧信号，然后采用上述频域分带方法直接将该语音信号转换成一个频域信号，并将其作为所述语音信号对应的频域信号，其中所述频域信号所在的频带包括至少两个频带区间，一个频带区间表征一个频点。例如，可以获取麦克风在一段较短时长内的采集的语音信号，比如麦克风在10ms内的所采集的语音信号，然后直接对这10ms的语音信号进行FTT处理，得到所述语音信号对应的频域信号，并将所述频域信号所在的频带划分为128频带区间，即128个频点。Optionally, in the specific implementation process of the present invention, the speech signal collected by the microphone may not be processed in frames, but the speech signal may be regarded as a frame signal, and then the speech signal may be directly converted into A frequency domain signal is used as a frequency domain signal corresponding to the speech signal, wherein the frequency band of the frequency domain signal includes at least two frequency band intervals, and one frequency band interval represents a frequency point. For example, it is possible to obtain the voice signal collected by the microphone within a short period of time, such as the voice signal collected by the microphone within 10 ms, and then directly perform FTT processing on the voice signal of 10 ms to obtain the frequency domain corresponding to the voice signal signal, and divide the frequency band where the frequency domain signal is located into 128 frequency band intervals, that is, 128 frequency points.
步骤120：计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点。Step 120: Calculate the first ratio between the first signal energy of the frequency domain signal at the first frequency point and the total energy of the frequency domain signal, and the first frequency point is the frequency domain where the frequency domain signal is located any frequency.
具体的，计算出第一频点所表征的频带区间内的信号的能量，即第一信号能量，将其除以所述第一频点所在频域信号的总能量，得到所述第一比值。由于此部分内容为本领域技术人员所熟知，所以此处就不再多做说明。Specifically, calculate the energy of the signal in the frequency band interval represented by the first frequency point, that is, the first signal energy, and divide it by the total energy of the frequency domain signal where the first frequency point is located to obtain the first ratio . Since this part of the content is well known to those skilled in the art, no further description is given here.
步骤130：根据扬声器的语音输入信号的能量确定啸叫门限值。Step 130: Determine the howling threshold according to the energy of the voice input signal from the speaker.
根据扬声器的语音输入信号的能量确定啸叫门限值，至少包括以下两种具体实现方式：Determining the howling threshold value according to the energy of the voice input signal of the loudspeaker includes at least the following two specific implementation methods:
第1种：根据扬声器的语音输入信号的平均能量所属的预设范围确定所述啸叫门限值。Type 1: The howling threshold is determined according to a preset range to which the average energy of the voice input signal of the loudspeaker belongs.
具体的，根据所述语音输入信号的能量和所述语音输入信号的时间长度计算出所述语音输入信号在单位时间内的平均能量，然后判断所述扬声器的语音输入信号的平均能量所属的预设范围：在所述扬声器的语音输入信号的平均能量属于第一预定范围时，确定所述啸叫门限值为第三啸叫门限值；在所述扬声器的语音输入信号的平均能量属于第二预定范围时，确定所述啸叫门限值为第四啸叫门限值。其中，所述第三啸叫门限值大于0且小于1，所述第二预定范围内的最小值大于等于0，所述第二预定范围内的最大值小于所述第一预定范围内的最小值，所述第四啸叫门限值大于所述第三啸叫门限值且小于1，所述预设范围的具体划分依实际情况而定，本发明实施例不做具体限制。Specifically, calculate the average energy of the voice input signal per unit time according to the energy of the voice input signal and the time length of the voice input signal, and then determine the preset to which the average energy of the voice input signal of the speaker belongs. Set the range: when the average energy of the voice input signal of the speaker belongs to the first predetermined range, determine that the howling threshold is the third howling threshold; when the average energy of the voice input signal of the speaker belongs to the second When the range is predetermined, it is determined that the howling threshold is a fourth howling threshold. Wherein, the third howling threshold value is greater than 0 and less than 1, the minimum value within the second predetermined range is greater than or equal to 0, and the maximum value within the second predetermined range is less than the minimum value within the first predetermined range. value, the fourth howling threshold is greater than the third howling threshold and less than 1, and the specific division of the preset range depends on actual conditions, and is not specifically limited in this embodiment of the present invention.
例如，参照图2，在所述扬声器的语音输入信号的平均能量属于第一预定范围[0，70]时，确定所述啸叫门限值为第三啸叫门限值，即0.5；在所述扬声器的语音输入信号的平均能量属于第二预定范围(70，+∞)时，确定所述啸叫门限值为第四啸叫门限值，即0.1。第三啸叫门限值和第四啸叫门限值的具体数值可以根据实际需要进行调整，本发明实施例不做具体限制。For example, referring to Fig. 2, when the average energy of the voice input signal of the speaker belongs to the first predetermined range [0, 70], it is determined that the howling threshold value is the third howling threshold value, namely 0.5; When the average energy of the voice input signal of the loudspeaker falls within the second predetermined range (70, +∞), determine the howling threshold as a fourth howling threshold, ie, 0.1. Specific values of the third howling threshold and the fourth howling threshold may be adjusted according to actual needs, and are not specifically limited in this embodiment of the present invention.
当然，在具体实施过程中，划分扬声器的语音输入信号的平均能量所属的预设范围除了上述划分形式外，还可以有其他形式，本发明实施例对此不做具体限制。比如：可以对第一预定范围和第二预定范围进行细分，确定出多个子范围，当所述扬声器的语音输入信号的平均能量属于不同的子范围时，确定出的啸叫门限值不同。Of course, in a specific implementation process, besides the above-mentioned division form, other forms may be used to divide the preset range to which the average energy of the speech input signal of the speaker belongs, which is not specifically limited in this embodiment of the present invention. For example, the first predetermined range and the second predetermined range may be subdivided to determine multiple sub-ranges, and when the average energy of the voice input signal of the speaker belongs to different sub-ranges, the determined howling thresholds are different.
例如，参照图3，当所述扬声器的语音输入信号的平均能量属于第一子范围[0，30]时，确定所述啸叫门限值为0.5；当所述扬声器的语音输入信号的平均能量属于第二子范围(30，100]时，确定所述啸叫门限值为0.35；当所述扬声器的语音输入信号的平均能量属于第三子范围(100，200]时，确定所述啸叫门限值为0.25；当所述扬声器的语音输入信号的平均能量属于第四子范围(200，+∞)时，确定所述啸叫门限值为0.15。For example, referring to Fig. 3, when the average energy of the voice input signal of the speaker belongs to the first sub-range [0,30], it is determined that the howling threshold value is 0.5; when the average energy of the voice input signal of the speaker is When belonging to the second sub-range (30, 100], determine the howling threshold value to be 0.35; when the average energy of the voice input signal of the loudspeaker belongs to the third sub-range (100, 200], determine the howling gate The limit value is 0.25; when the average energy of the voice input signal of the loudspeaker belongs to the fourth subrange (200, +∞), the howling threshold value is determined to be 0.15.
下面介绍一种划分扬声器的语音输入信号的平均能量所属范围的特例：根据扬声器是否有语音输入确定所述啸叫门限值。A special example of dividing the range of the average energy of the voice input signal of the loudspeaker is introduced below: the howling threshold is determined according to whether the loudspeaker has voice input.
具体的，判断所述扬声器的语音输入信号的平均能量是否大于0；若为是，则确定所述扬声器有语音输入，并确定所述啸叫门限值为第一啸叫门限值；若为否，则确定所述扬声器无语音输入，并确定所述啸叫门限值为第二啸叫门限值；所述第二啸叫门限值大于所述第一啸叫门限值且小于1，所述第一啸叫门限值大于0且小于1。当然，也可以直接对所述扬声器的语音输入信号的能量进行判断，在所述扬声器的语音输入信号的能量大于0时，确定所述啸叫门限值为所述第一啸叫门限值；在所述扬声器的语音输入信号的能量不大于0时，确定所述啸叫门限值为所述第二啸叫门限值。Specifically, determine whether the average energy of the voice input signal of the speaker is greater than 0; if yes, then determine that the speaker has voice input, and determine that the howling threshold is the first howling threshold; if not , then it is determined that the speaker has no voice input, and it is determined that the howling threshold is a second howling threshold; the second howling threshold is greater than the first howling threshold and less than 1, and the first howling threshold is A howling threshold is greater than 0 and less than 1. Of course, it is also possible to directly judge the energy of the voice input signal of the speaker, and when the energy of the voice input signal of the speaker is greater than 0, determine that the howling threshold is the first howling threshold; When the energy of the voice input signal of the speaker is not greater than 0, determine that the howling threshold is the second howling threshold.
例如，在所述扬声器有语音输入时，确定所述啸叫门限值为第一啸叫门限值0.1，在所述扬声器无语音输入时，确定所述啸叫门限值为第二啸叫门限值0.14。第一啸叫门限值和第二啸叫门限值的具体数值可以根据实际需要进行调整，本发明实施例不做具体限制。For example, when the speaker has voice input, determine that the howling threshold is the first howling threshold 0.1, and when the speaker has no voice input, determine that the howling threshold is the second howling threshold 0.14. Specific values of the first howling threshold and the second howling threshold may be adjusted according to actual needs, and are not specifically limited in this embodiment of the present invention.
第2种：基于公式y＝f(x)确定所述啸叫门限值。Type 2: determining the howling threshold based on the formula y=f(x).
具体的，y为所述啸叫门限值，x为所述扬声器的语音输入信号的平均能量，x等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，f(x)表示以x为变量的函数，f(x)大于0且小于1，所述啸叫门限值y与单位时间内所述扬声器的语音输入信号的能量x呈负相关关系。所述扬声器的语音输入信号的平均能量越大，表示发生啸叫的可能性就越大，所述啸叫门限值就越小。Specifically, y is the howling threshold, x is the average energy of the voice input signal of the speaker, x is equal to the energy of the voice input signal of the speaker divided by the duration of the voice input signal of the speaker, f( x) represents a function with x as a variable, f(x) is greater than 0 and less than 1, and the howling threshold y is negatively correlated with the energy x of the voice input signal of the loudspeaker per unit time. The greater the average energy of the voice input signal of the loudspeaker, the greater the possibility of howling, and the smaller the howling threshold.
例如，参照图4，图中x表征扬声器的语音输入信号的平均能量，y表征啸叫门限值。随着扬声器的语音输入信号的平均能量的增大，所述啸叫门限值y＝f(x)的值越小。比如当x＝10时，y＝0.7；当x＝50时，y＝0.5；当x＝150时，y＝0.12。For example, referring to FIG. 4 , x in the figure represents the average energy of the voice input signal of the speaker, and y represents the howling threshold. As the average energy of the voice input signal of the loudspeaker increases, the value of the howling threshold y=f(x) becomes smaller. For example, when x=10, y=0.7; when x=50, y=0.5; when x=150, y=0.12.
可选的，在采用以上两种方式确定啸叫门限值时，为了规避偶然的噪音引起的扬声器的语音输入信号的平均能量较大的情况，使确定出的啸叫门限值更为准确，需要限定所述扬声器的语音输入信号为所述扬声器在超过某一阈值的时长范围内接收的语音输入信号。例如，所述扬声器的语音输入信号为所述扬声器在超过5ms时长范围内所接收的语音输入信号，或者是所述扬声器在超过20ms时长范围内所接收的语音输入信号，又或者是所述扬声器在超过50ms时长范围内所接收的语音输入信号，本发明实施例对第一阈值不做具体限制。Optionally, when using the above two methods to determine the howling threshold value, in order to avoid the situation that the average energy of the voice input signal of the loudspeaker caused by accidental noise is relatively large, and to make the determined howling threshold value more accurate, it is necessary to The voice input signal of the loudspeaker is defined as the voice input signal received by the loudspeaker within a time range exceeding a certain threshold. For example, the voice input signal of the speaker is a voice input signal received by the speaker within a duration exceeding 5 ms, or a voice input signal received by the speaker within a duration exceeding 20 ms, or the speaker For voice input signals received within a duration exceeding 50 ms, the embodiment of the present invention does not specifically limit the first threshold.
步骤140：判断所述第一比值是否大于所述啸叫门限值。Step 140: Determine whether the first ratio is greater than the howling threshold.
具体的，将所述频域信号在第一频点所对应的第一比值与确定出的啸叫门限值进行大小比较。若所述第一比值小于或等于所述啸叫门限值，则确定所述第一频点不是啸叫频点；若所述第一比值大于所述啸叫门限值，则执行步骤150：Specifically, the first ratio corresponding to the frequency domain signal at the first frequency point is compared with the determined howling threshold value. If the first ratio is less than or equal to the howling threshold, it is determined that the first frequency point is not a howling frequency point; if the first ratio is greater than the howling threshold value, perform step 150:
步骤150：在所述第一比值大于所述啸叫门限值时，确定所述第一频点为啸叫频点。Step 150: When the first ratio is greater than the howling threshold, determine that the first frequency is a howling frequency.
可选的，在确定所述第一频点为啸叫频点之后，所述方法还可以包括：Optionally, after determining that the first frequency point is a howling frequency point, the method may further include:
步骤160：对所述第一频点进行啸叫抑制。Step 160: Perform howling suppression on the first frequency point.
具体的，对第一频点进行啸叫抑制可以有多种实现方式。例如，可以采用陷波滤波器或带阻滤波器对该信号进行衰减处理，实现啸叫抑制；又例如，可以将第一频点对应的信号进行频移，使得第一频点对应的信号每一次回到麦克风时的频率都不一样，进而使得信号无法叠加，实现啸叫抑制。当然，在具体实施过程中还可以有其他啸叫抑制的实现方法，由于该部分内容为本领域技术人员所熟知，所以此处就不再多做举例说明。Specifically, there may be multiple implementation manners for suppressing howling at the first frequency point. For example, the signal can be attenuated by using a notch filter or a band-stop filter to suppress howling; The frequency is different when returning to the microphone at a time, so that the signals cannot be superimposed and the howling suppression is realized. Certainly, there may be other implementation methods of howling suppression in the specific implementation process, and since this part of the content is well known to those skilled in the art, no more examples will be given here.
在上述方案中，通过对麦克风采集的语音信号进行频域处理得到麦克风采集的语音信号在每个频点处的能量与语音信号总能量的比值，通过对扬声器的语音输入信号的检测和分析确定出啸叫门限值，然后将麦克风采集的语音信号在每个频点所对应的能量比值与该啸叫门限值进行比较，在比值大于该啸叫门限值时将该比值对应的频点确定为啸叫频点，并对啸叫频点进行啸叫抑制。本方案根据麦克风采集的语音信号和扬声器的语音输入信号确定啸叫频点，能够使得检测出的啸叫频点更加准确，进而使得对啸叫的抑制效果更好，提升了扩声系统的性能。In the above scheme, the ratio of the energy of the voice signal collected by the microphone at each frequency point to the total energy of the voice signal is obtained by performing frequency domain processing on the voice signal collected by the microphone, and determined by detecting and analyzing the voice input signal of the speaker Then compare the energy ratio corresponding to each frequency point of the voice signal collected by the microphone with the howling threshold value, and determine the frequency point corresponding to the ratio when the ratio is greater than the howling threshold value as Howling frequency point, and suppress howling frequency point. This solution determines the howling frequency point based on the voice signal collected by the microphone and the voice input signal of the speaker, which can make the detected howling frequency point more accurate, thereby making the howling suppression effect better and improving the performance of the sound reinforcement system .
不仅如此，上述方案还提供了多种确定啸叫门限值的方法，例如根据扬声器的语音输入信号的平均能量的所属范围确定啸叫门限值或者根据公式y＝f(x)确定所述啸叫门限值，在啸叫发生概率较大的情况下减小啸叫门限值以及在啸叫发生概率较小的情况下增大啸叫门限值，实现了对啸叫门限值的灵活调整，进一步提高了对啸叫检测的准确性。Not only that, the above solution also provides a variety of methods for determining the howling threshold, for example, determining the howling threshold according to the range of the average energy of the voice input signal of the speaker or determining the howling threshold according to the formula y=f(x) Limit value, reduce the howling threshold value when the howling probability is high and increase the howling threshold value when the howling probability is small, realize the flexible adjustment of the howling threshold value, and further improve The accuracy of howling detection is improved.
实施例二Embodiment two
本发明实施例二提供一种啸叫检测装置，用于实现上述实施例一所提供的啸叫检测方法。参照图5，所述装置包括：Embodiment 2 of the present invention provides a howling detection device, which is used to implement the howling detection method provided in Embodiment 1 above. With reference to Fig. 5, described device comprises:
转换单元201，用于将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；A conversion unit 201, configured to convert the voice signal collected by the microphone from the time domain to the frequency domain, to obtain a frequency domain signal corresponding to the voice signal;
计算单元202，用于计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；A calculation unit 202, configured to calculate a first ratio between the first signal energy of the frequency domain signal at a first frequency point and the total energy of the frequency domain signal, where the first frequency point is where the frequency domain signal is located Any frequency point in the frequency domain;
第一确定单元203，用于根据扬声器的语音输入信号的能量确定啸叫门限值；The first determination unit 203 is configured to determine the howling threshold value according to the energy of the voice input signal of the speaker;
判断单元204，用于判断所述第一比值是否大于所述啸叫门限值；A judging unit 204, configured to judge whether the first ratio is greater than the howling threshold;
第二确定单元205，用于在所述第一比值大于所述啸叫门限值时，确定所述第一频点为啸叫频点。The second determining unit 205 is configured to determine that the first frequency point is a howling frequency point when the first ratio is greater than the howling threshold value.
可选的，所述第一确定单元203用于判断所述扬声器的语音输入信号的能量是否大于0；在所述扬声器的语音输入信号的能量大于0时确定所述扬声器有语音输入，并确定所述啸叫门限值为第一啸叫门限值，所述第一啸叫门限值大于0且小于1；在所述扬声器的语音输入信号的能量不大于0时确定所述扬声器无语音输入，并确定所述啸叫门限值为第二啸叫门限值，所述第二啸叫门限值大于所述第一啸叫门限值且小于1。Optionally, the first determination unit 203 is configured to determine whether the energy of the voice input signal of the speaker is greater than 0; when the energy of the voice input signal of the speaker is greater than 0, determine that the speaker has voice input, and determine The howling threshold value is a first howling threshold value, and the first howling threshold value is greater than 0 and less than 1; when the energy of the voice input signal of the speaker is not greater than 0, it is determined that the speaker has no voice input, And determine that the howling threshold is a second howling threshold, where the second howling threshold is greater than the first howling threshold and less than 1.
可选的，所述第一确定单元203用于在所述扬声器的语音输入信号的平均能量属于第一预定范围时，确定所述啸叫门限值为第三啸叫门限值，所述扬声器的语音输入信号的平均能量等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，所述第三啸叫门限值大于0且小于1；在所述扬声器的语音输入信号的平均能量属于第二预定范围时，确定所述啸叫门限值为第四啸叫门限值，所述第二预定范围内的最小值大于等于0，所述第二预定范围内的最大值小于所述第一预定范围内的最小值，所述第四啸叫门限值大于所述第三啸叫门限值且小于1。Optionally, the first determining unit 203 is configured to determine that the howling threshold is a third howling threshold when the average energy of the voice input signal of the speaker falls within a first predetermined range, and the howling threshold of the speaker is The average energy of the voice input signal is equal to the energy of the voice input signal of the loudspeaker divided by the duration of the voice input signal of the loudspeaker, and the third howling threshold is greater than 0 and less than 1; When the average energy of the value belongs to the second predetermined range, it is determined that the howling threshold value is the fourth howling threshold value, the minimum value in the second predetermined range is greater than or equal to 0, and the maximum value in the second predetermined range is less than The minimum value within the first predetermined range, the fourth howling threshold is greater than the third howling threshold and less than 1.
可选的，所述第一确定单元203用于基于公式：y＝f(x)确定所述啸叫门限值；其中，y为所述啸叫门限值，x为所述扬声器的语音输入信号的平均能量，x等于所述扬声器的语音输入信号的能量除以所述扬声器的语音输入信号的时长，f(x)表示以x为变量的函数，f(x)大于0且小于1，所述啸叫门限值y与单位时间内所述扬声器的语音输入信号的能量x呈负相关关系。Optionally, the first determining unit 203 is configured to determine the howling threshold based on the formula: y=f(x); wherein, y is the howling threshold, and x is the voice input signal of the speaker The average energy of x is equal to the energy of the speech input signal of the loudspeaker divided by the duration of the speech input signal of the loudspeaker, f(x) represents a function with x as a variable, f(x) is greater than 0 and less than 1, so The howling threshold value y is negatively correlated with the energy x of the voice input signal of the loudspeaker per unit time.
可选的，所述扬声器的语音输入信号为所述扬声器在超过第一阈值的时长范围内接收的语音输入信号。Optionally, the voice input signal of the speaker is a voice input signal received by the speaker within a time range exceeding a first threshold.
可选的，所述转换单元201用于将所述麦克风采集的语音信号进行分帧处理，得到M帧时域信号，M为大于等于1的正整数；对所述M帧时域信号中的任意一帧时域信号进行快速傅里叶变换FTT处理，得到所述频域信号。Optionally, the conversion unit 201 is configured to process the voice signal collected by the microphone into frames to obtain M frames of time domain signals, where M is a positive integer greater than or equal to 1; for the M frames of time domain signals The time-domain signal of any frame is processed by fast Fourier transform (FTT) to obtain the frequency-domain signal.
以上各单元所执行操作步骤的具体实现方式可以参照实施例一中各对应步骤的具体实现方式，本发明实施例不再赘述。For the specific implementation manner of the operation steps performed by the above units, reference may be made to the specific implementation manner of each corresponding step in Embodiment 1, which will not be repeated in this embodiment of the present invention.
实施例三Embodiment three
本发明实施例三提供一种啸叫检测系统，用于实现上述实施例一所提供的啸叫检测方法。参照图6，所述系统包括拾音电路301、信号处理电路302以及信号判断电路303.Embodiment 3 of the present invention provides a howling detection system, which is used to implement the howling detection method provided in Embodiment 1 above. With reference to Fig. 6, described system comprises pickup circuit 301, signal processing circuit 302 and signal judging circuit 303.
所述拾音电路301与扩声系统的麦克风和扬声器分别连接，用于获取麦克风采集的语音信号和扬声器的语音输入信号；The sound pickup circuit 301 is respectively connected with the microphone and the loudspeaker of the sound reinforcement system, and is used to obtain the voice signal collected by the microphone and the voice input signal of the loudspeaker;
所述信号处理电路302与所述拾音电路301相连，用于将麦克风采集的语音信号从时域变换到频域，得到所述语音信号对应的频域信号；The signal processing circuit 302 is connected to the sound pickup circuit 301, and is used to convert the voice signal collected by the microphone from the time domain to the frequency domain, so as to obtain a frequency domain signal corresponding to the voice signal;
所述信号判断电路303与所述信号处理电路302相连，用于计算所述频域信号在第一频点处的第一信号能量与所述频域信号总能量间的第一比值，所述第一频点为所述频域信号所在频域内的任一频点；根据扬声器的语音输入信号的能量确定啸叫门限值；判断所述第一比值是否大于所述啸叫门限值；在所述第一比值大于所述啸叫门限值时，确定所述第一频点为啸叫频点。The signal judging circuit 303 is connected to the signal processing circuit 302, and is used to calculate a first ratio between the first signal energy of the frequency domain signal at the first frequency point and the total energy of the frequency domain signal, the The first frequency point is any frequency point in the frequency domain where the frequency domain signal is located; determine the howling threshold value according to the energy of the voice input signal of the loudspeaker; judge whether the first ratio is greater than the howling threshold value; When the first ratio is greater than the howling threshold, it is determined that the first frequency point is a howling frequency point.
在具体实施过程中，所述该啸叫检测系统可以全部集成在扩声系统内，也可以部分集成在扩声系统内，还可以单独设置在扩声系统外并与扩声系统保持连接，本发明实施例对此不做具体限制。In the specific implementation process, the howling detection system can be fully integrated in the sound reinforcement system, can also be partly integrated in the sound reinforcement system, and can also be separately installed outside the sound reinforcement system and kept connected with the sound reinforcement system. The embodiment of the invention does not specifically limit this.
例如，当麦克风和扬声器分开设置在两个不同设备上时，可以将拾音电路301、信号处理电路302和信号判断电路303全部集成在麦克风内，或者全部集成在扬声器内，又或者将拾音电路301集成在麦克风内而将信号处理电路302和信号判断电路303集成在扬声器内；又例如，当麦克风和扬声器为同一扩声设备的不同组成部件时，可以将拾音电路301、信号处理电路302和信号判断电路303全部集成在所述扩声设备内，或者将拾音电路301、信号处理电路302和信号判断电路303中的部分集成在在所述扩声设备内，又或者将拾音电路301、信号处理电路302和信号判断电路303全部设置在所述扩声设备外，并将所述拾音电路与所述扩声设备相连。For example, when the microphone and the speaker are separately installed on two different devices, the sound pickup circuit 301, the signal processing circuit 302 and the signal judgment circuit 303 can all be integrated in the microphone, or all integrated in the speaker, or the sound pickup circuit 301 can be integrated into the speaker. The circuit 301 is integrated in the microphone and the signal processing circuit 302 and the signal judging circuit 303 are integrated in the speaker; 302 and signal judging circuit 303 are all integrated in the sound reinforcement equipment, or part of the sound pickup circuit 301, signal processing circuit 302 and signal judgment circuit 303 are integrated in the sound reinforcement equipment, or the sound pickup The circuit 301, the signal processing circuit 302 and the signal judging circuit 303 are all arranged outside the sound reinforcement equipment, and the sound pickup circuit is connected with the sound reinforcement equipment.
其中，所述拾音电路301、所述信号处理电路302以及所述信号判断电路303的具体实现方式可以为集成电路，或集成芯片，或集成电路与集成芯片的组合等，本发明实施例不做具体限制。例如，信号处理电路302可以通过模数转换器(Analog to DigitalConverter，ADC)实现；所述信号判断电路303可以通过现场可编程门阵列芯片(Field－Programmable Gate Array，FPGA)或者微处理器(Digital Signal Processor，DSP)实现。Wherein, the specific implementation manner of the pickup circuit 301, the signal processing circuit 302 and the signal judgment circuit 303 may be an integrated circuit, or an integrated chip, or a combination of an integrated circuit and an integrated chip, etc., the embodiment of the present invention does not Make specific restrictions. For example, the signal processing circuit 302 can be implemented by an analog-to-digital converter (Analog to Digital Converter, ADC); the signal judgment circuit 303 can be implemented by a field programmable gate array chip (Field-Programmable Gate Array, FPGA) or a microprocessor (Digital Signal Processor, DSP) implementation.
以上所述拾音电路301、所述信号处理电路302以及所述信号判断电路303所执行的操作步骤的具体实现方式可以参照实施例一中各对应步骤的具体实现方式，本发明实施例不再赘述。The specific implementation of the operation steps performed by the sound pickup circuit 301, the signal processing circuit 302, and the signal judgment circuit 303 above can refer to the specific implementation of each corresponding step in Embodiment 1, and the embodiment of the present invention is no longer repeat.
实施例四Embodiment four
本发明实施例四提供一种啸叫检测设备，参照图7，所述设备包括：Embodiment 4 of the present invention provides a howling detection device. Referring to FIG. 7, the device includes:
至少一个处理器401；以及，at least one processor 401; and,
与所述至少一个处理器401连接的存储器402；其中，A memory 402 connected to the at least one processor 401; wherein,
所述存储器402存储有可被所述至少一个处理器401执行的指令，所述指令被所述至少一个处理器401执行，以使所述至少一个处理器401能够执行上述实施例一所述的啸叫检测方法。The memory 402 stores instructions that can be executed by the at least one processor 401, and the instructions are executed by the at least one processor 401, so that the at least one processor 401 can execute the method described in the first embodiment above. Howling detection method.
以上处理器401所执行的操作步骤的具体实现方式可以参照实施例一中各对应步骤的具体实现方式，本发明实施例不再赘述。For the specific implementation manner of the above operation steps performed by the processor 401, reference may be made to the specific implementation manner of each corresponding step in Embodiment 1, and details will not be repeated in this embodiment of the present invention.
需要说明的是，以上处理器可以是一个处理元件，也可以是多个处理元件的统称。例如，该处理器可以是中央处理器(Central Processing Unit，CPU)，也可以是特定集成电路(Application Specific Integrated Circuit，ASIC)，或者是被配置成实施本发明实施例的一个或多个集成电路，例如：一个或多个微处理器(Digital Signal Processor，DSP)，或，一个或者多个现场可编程门阵列(Field Programmable Gate Array，FPGA)。It should be noted that the above processor may be one processing element, or may be a general term for multiple processing elements. For example, the processor may be a central processing unit (Central Processing Unit, CPU), or a specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention , for example: one or more microprocessors (Digital Signal Processor, DSP), or, one or more field programmable gate arrays (Field Programmable Gate Array, FPGA).
存储器可以是一个存储装置，也可以是多个存储元件的统称，且用于存储可执行程序代码、数据等。且存储器可以包括随机存储器(Random-Access Memory，RAM)，也可以包括非易失性存储器(Non-Volatile Memory，NVM)，例如磁盘存储器，闪存(Flash)等。The memory may be a storage device, or a general term for multiple storage elements, and is used to store executable program codes, data, and the like. And the memory may include random-access memory (Random-Access Memory, RAM), and may also include non-volatile memory (Non-Volatile Memory, NVM), such as disk memory, flash memory (Flash), and the like.
本发明实施例中提供的一个或多个技术方案，至少具有如下技术效果或优点：One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:
1、通过对麦克风采集的语音信号进行频谱处理得到麦克风采集的语音信号在每个频点处的能量与语音信号总能量的比值，通过对扬声器的语音输入信号的检测和分析确定出啸叫门限值，然后将麦克风采集的语音信号在每个频点所对应的能量比值与该啸叫门限值进行比较，在比值大于该啸叫门限值时将该比值对应的频点确定为啸叫频点，并对啸叫频点进行啸叫抑制。本方案根据麦克风采集的语音信号和扬声器的语音输入信号确定啸叫频点，能够使得检测出的啸叫频点更加准确，进而使得对啸叫的抑制效果更好，提升了扩声系统的性能。1. The ratio of the energy of the voice signal collected by the microphone at each frequency point to the total energy of the voice signal is obtained by performing spectrum processing on the voice signal collected by the microphone, and the howling threshold is determined by detecting and analyzing the voice input signal of the speaker value, and then compare the energy ratio corresponding to each frequency point of the voice signal collected by the microphone with the howling threshold value, and determine the frequency point corresponding to the ratio as the howling frequency point when the ratio is greater than the howling threshold value , and perform howling suppression on howling frequency points. This solution determines the howling frequency point based on the voice signal collected by the microphone and the voice input signal of the speaker, which can make the detected howling frequency point more accurate, thereby making the howling suppression effect better and improving the performance of the sound reinforcement system .
2、提供了多种确定啸叫门限值的方法，例如根据扬声器的语音输入信号的平均能量的所属范围确定啸叫门限值或者根据公式y＝f(x)确定所述啸叫门限值，在啸叫发生概率较大的情况下减小啸叫门限值以及在啸叫发生概率较小的情况下增大啸叫门限值，实现了对啸叫门限值的灵活调整，进一步提高了对啸叫检测的准确性。2. A variety of methods for determining the howling threshold are provided, for example, determining the howling threshold according to the range of the average energy of the voice input signal of the loudspeaker or determining the howling threshold according to the formula y=f(x). The howling threshold value is reduced when the howling probability is high and the howling threshold value is increased when the howling probability is small, which realizes the flexible adjustment of the howling threshold value and further improves the howling threshold value. detection accuracy.
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer program products. Accordingly, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.
Claims (14)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710468828.3A CN109102819A (en) | 2017-06-20 | 2017-06-20 | One kind is uttered long and high-pitched sounds detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710468828.3A CN109102819A (en) | 2017-06-20 | 2017-06-20 | One kind is uttered long and high-pitched sounds detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109102819A true CN109102819A (en) | 2018-12-28 |
Family
ID=64795484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710468828.3A Pending CN109102819A (en) | 2017-06-20 | 2017-06-20 | One kind is uttered long and high-pitched sounds detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109102819A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110493693A (en) * | 2019-09-02 | 2019-11-22 | 昆腾微电子股份有限公司 | A kind of acoustic signal processing method and device |
| CN113223554A (en) * | 2021-03-15 | 2021-08-06 | 百度在线网络技术（北京）有限公司 | Wind noise detection method, device, equipment and storage medium |
| CN113225657A (en) * | 2021-04-16 | 2021-08-06 | 深圳木芯科技有限公司 | Multi-channel squeal suppression method based on double-microphone architecture |
| CN113271386A (en) * | 2021-05-14 | 2021-08-17 | 杭州朗和科技有限公司 | Howling detection method and device, storage medium and electronic equipment |
| CN113299304A (en) * | 2021-05-11 | 2021-08-24 | 泰凌微电子(上海)股份有限公司 | Method and device for inhibiting microphone howling and microphone |
| CN113316075A (en) * | 2021-05-11 | 2021-08-27 | 紫光展锐(重庆)科技有限公司 | Howling detection method and device and electronic equipment |
| CN114387985A (en) * | 2021-09-30 | 2022-04-22 | 长沙幻音电子科技有限公司 | Howling suppression method and system for stringed musical instrument and stringed musical instrument |
| CN114724576A (en) * | 2022-06-09 | 2022-07-08 | 广州市保伦电子有限公司 | Method, device and system for updating threshold in howling detection in real time |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1558551A (en) * | 2004-01-17 | 2004-12-29 | 崔光旭 | Base frequency shifting feedback controller |
| US20050220313A1 (en) * | 2004-03-30 | 2005-10-06 | Yamaha Corporation | Howling frequency component emphasis method and apparatus |
| CN1767695A (en) * | 2005-09-08 | 2006-05-03 | 南京大学 | Howling suppression method without loss of voice quality |
| CN201336660Y (en) * | 2008-04-30 | 2009-10-28 | 沈生祥 | Multifunctional teaching microphone based on 2.4G digital wireless frequency modulation and shift |
| CN101808260A (en) * | 2009-04-15 | 2010-08-18 | 赵凡 | Audio dynamic feedback suppression method |
| CN103338419A (en) * | 2013-06-29 | 2013-10-02 | 青岛歌尔声学科技有限公司 | Method and device for removing headset scream |
| CN103871418A (en) * | 2014-03-06 | 2014-06-18 | 北京飞利信电子技术有限公司 | Method and device for detecting howling frequency point of acoustic amplification system |
| CN104796822A (en) * | 2015-01-16 | 2015-07-22 | 北京中电兴发科技有限公司 | Audio howling detection method, video monitoring method and system using same |
| CN105118522A (en) * | 2015-08-27 | 2015-12-02 | 广州市百果园网络科技有限公司 | Noise detection method and device |
| CN105812993A (en) * | 2014-12-29 | 2016-07-27 | 联芯科技有限公司 | Howling detection and suppression method and device |
| CN105872910A (en) * | 2016-03-23 | 2016-08-17 | 成都普创通信技术股份有限公司 | Audio signal squeaking detection method |
| CN105895115A (en) * | 2016-04-01 | 2016-08-24 | 北京小米移动软件有限公司 | Squeal determining method and squeal determining device |
| US20160277833A1 (en) * | 2013-09-18 | 2016-09-22 | Zte Corporation | Piezoelectric Speaker Driving Device |
| CN105979197A (en) * | 2016-07-19 | 2016-09-28 | 块互动（北京）科技有限公司 | Remote conference control method and device based on automatic recognition of howling sound |
| CN106384597A (en) * | 2016-08-31 | 2017-02-08 | 广州市百果园网络科技有限公司 | Audio frequency data processing method and device |
| CN106486133A (en) * | 2015-08-27 | 2017-03-08 | 成都鼎桥通信技术有限公司 | One kind is uttered long and high-pitched sounds scene recognition method and equipment |
| CN106488052A (en) * | 2015-08-27 | 2017-03-08 | 成都鼎桥通信技术有限公司 | One kind is uttered long and high-pitched sounds scene recognition method and equipment |
-
2017
- 2017-06-20 CN CN201710468828.3A patent/CN109102819A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1558551A (en) * | 2004-01-17 | 2004-12-29 | 崔光旭 | Base frequency shifting feedback controller |
| US20050220313A1 (en) * | 2004-03-30 | 2005-10-06 | Yamaha Corporation | Howling frequency component emphasis method and apparatus |
| CN1767695A (en) * | 2005-09-08 | 2006-05-03 | 南京大学 | Howling suppression method without loss of voice quality |
| CN201336660Y (en) * | 2008-04-30 | 2009-10-28 | 沈生祥 | Multifunctional teaching microphone based on 2.4G digital wireless frequency modulation and shift |
| CN101808260A (en) * | 2009-04-15 | 2010-08-18 | 赵凡 | Audio dynamic feedback suppression method |
| CN103338419A (en) * | 2013-06-29 | 2013-10-02 | 青岛歌尔声学科技有限公司 | Method and device for removing headset scream |
| US20160277833A1 (en) * | 2013-09-18 | 2016-09-22 | Zte Corporation | Piezoelectric Speaker Driving Device |
| CN103871418A (en) * | 2014-03-06 | 2014-06-18 | 北京飞利信电子技术有限公司 | Method and device for detecting howling frequency point of acoustic amplification system |
| CN105812993A (en) * | 2014-12-29 | 2016-07-27 | 联芯科技有限公司 | Howling detection and suppression method and device |
| CN104796822A (en) * | 2015-01-16 | 2015-07-22 | 北京中电兴发科技有限公司 | Audio howling detection method, video monitoring method and system using same |
| CN105118522A (en) * | 2015-08-27 | 2015-12-02 | 广州市百果园网络科技有限公司 | Noise detection method and device |
| CN106486133A (en) * | 2015-08-27 | 2017-03-08 | 成都鼎桥通信技术有限公司 | One kind is uttered long and high-pitched sounds scene recognition method and equipment |
| CN106488052A (en) * | 2015-08-27 | 2017-03-08 | 成都鼎桥通信技术有限公司 | One kind is uttered long and high-pitched sounds scene recognition method and equipment |
| CN105872910A (en) * | 2016-03-23 | 2016-08-17 | 成都普创通信技术股份有限公司 | Audio signal squeaking detection method |
| CN105895115A (en) * | 2016-04-01 | 2016-08-24 | 北京小米移动软件有限公司 | Squeal determining method and squeal determining device |
| CN105979197A (en) * | 2016-07-19 | 2016-09-28 | 块互动（北京）科技有限公司 | Remote conference control method and device based on automatic recognition of howling sound |
| CN106384597A (en) * | 2016-08-31 | 2017-02-08 | 广州市百果园网络科技有限公司 | Audio frequency data processing method and device |
Non-Patent Citations (1)
| Title |
|---|
| 梁瑞宇等: "啸叫快速抑制的助听器回声抵消算法", 《声学学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110493693A (en) * | 2019-09-02 | 2019-11-22 | 昆腾微电子股份有限公司 | A kind of acoustic signal processing method and device |
| CN113223554A (en) * | 2021-03-15 | 2021-08-06 | 百度在线网络技术（北京）有限公司 | Wind noise detection method, device, equipment and storage medium |
| CN113225657A (en) * | 2021-04-16 | 2021-08-06 | 深圳木芯科技有限公司 | Multi-channel squeal suppression method based on double-microphone architecture |
| CN113299304A (en) * | 2021-05-11 | 2021-08-24 | 泰凌微电子(上海)股份有限公司 | Method and device for inhibiting microphone howling and microphone |
| CN113316075A (en) * | 2021-05-11 | 2021-08-27 | 紫光展锐(重庆)科技有限公司 | Howling detection method and device and electronic equipment |
| CN113316075B (en) * | 2021-05-11 | 2023-02-14 | 紫光展锐(重庆)科技有限公司 | Howling detection method and device and electronic equipment |
| CN113299304B (en) * | 2021-05-11 | 2023-09-12 | 泰凌微电子(上海)股份有限公司 | Method and device for suppressing microphone howling and microphone |
| CN113271386A (en) * | 2021-05-14 | 2021-08-17 | 杭州朗和科技有限公司 | Howling detection method and device, storage medium and electronic equipment |
| CN114387985A (en) * | 2021-09-30 | 2022-04-22 | 长沙幻音电子科技有限公司 | Howling suppression method and system for stringed musical instrument and stringed musical instrument |
| CN114724576A (en) * | 2022-06-09 | 2022-07-08 | 广州市保伦电子有限公司 | Method, device and system for updating threshold in howling detection in real time |
| CN114724576B (en) * | 2022-06-09 | 2022-10-04 | 广州市保伦电子有限公司 | Method, device and system for updating threshold in howling detection in real time |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109102819A (en) | One kind is uttered long and high-pitched sounds detection method and device | |
| CN111149370B (en) | Howling detection in a conferencing system | |
| EP2773137A2 (en) | Microphone sensitivity difference correction device | |
| CN109074814B (en) | Noise detection method and terminal equipment | |
| CN108305637B (en) | Earphone voice processing method, terminal equipment and storage medium | |
| JP2016529555A (en) | Voice activity detection method and apparatus | |
| US20150162021A1 (en) | Spectral Comb Voice Activity Detection | |
| JP2010112996A (en) | Voice processing device, voice processing method and program | |
| CN105469807B (en) | A kind of more fundamental frequency extracting methods and device | |
| WO2018014673A1 (en) | Method and device for howling detection | |
| WO2020228107A1 (en) | Audio repair method and device, and readable storage medium | |
| CN104464752A (en) | Sound feedback detection method and device | |
| US20240386051A1 (en) | Methods and Apparatus to Fingerprint an Audio Signal Via Exponential Normalization | |
| US10070219B2 (en) | Sound feedback detection method and device | |
| JP2016042613A (en) | Target speech section detector, target speech section detection method, target speech section detection program, audio signal processing device and server | |
| CN110556128B (en) | Voice activity detection method and device and computer readable storage medium | |
| TW201503116A (en) | Method for using voiceprint identification to operate voice recoginition and electronic device thereof | |
| CN110148421B (en) | Residual echo detection method, terminal and device | |
| CN113316075B (en) | Howling detection method and device and electronic equipment | |
| CN106847299B (en) | Time delay estimation method and device | |
| JP5772591B2 (en) | Audio signal processing device | |
| EP3057344B1 (en) | Sound field measurement device, sound field measurement method, and sound field measurement program | |
| JP6314475B2 (en) | Audio signal processing apparatus and program | |
| CN111477246A (en) | Voice processing method and device and intelligent terminal | |
| CN112017649B (en) | Audio processing method, device, electronic device and readable storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181228 |