CN116312597A - Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit - Google Patents
Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit Download PDFInfo
- Publication number
- CN116312597A CN116312597A CN202310273614.6A CN202310273614A CN116312597A CN 116312597 A CN116312597 A CN 116312597A CN 202310273614 A CN202310273614 A CN 202310273614A CN 116312597 A CN116312597 A CN 116312597A
- Authority
- CN
- China
- Prior art keywords
- howling
- peak
- power ratio
- frequency point
- suspected
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012549 training Methods 0.000 claims abstract description 46
- 230000003044 adaptive effect Effects 0.000 claims abstract description 37
- 238000001514 detection method Methods 0.000 claims abstract description 33
- 238000001228 spectrum Methods 0.000 claims abstract description 31
- 238000011068 loading method Methods 0.000 claims abstract description 12
- 230000014759 maintenance of location Effects 0.000 claims abstract description 12
- 238000001914 filtration Methods 0.000 claims abstract description 9
- 230000001629 suppression Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 9
- 230000004044 response Effects 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 5
- 238000007796 conventional method Methods 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- CRBHXDCYXIISFC-UHFFFAOYSA-N 2-(Trimethylammonio)ethanolate Chemical compound C[N+](C)(C)CC[O-] CRBHXDCYXIISFC-UHFFFAOYSA-N 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000003071 parasitic effect Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
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
- G10L21/0264—Noise filtering characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R3/00—Circuits for transducers, loudspeakers or microphones
- H04R3/02—Circuits for transducers, loudspeakers or microphones for preventing acoustic reaction, i.e. acoustic oscillatory feedback
-
- 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
- G10L2021/02082—Noise filtering the noise being echo, reverberation of the speech
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/40—Arrangements for reducing harmonics
Landscapes
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Otolaryngology (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention discloses a method, a device, equipment and a system for detecting and suppressing howling of an aircraft cockpit, and belongs to the technical field of avionics. The method of the invention comprises the following steps: the peak-to-average power ratio, the peak amplitude retention between frames, the peak harmonic power ratio and the slope deviation of the amplitude spectrum between frames are combined to detect howling on the voice signal; after the occurrence of howling is detected, loading a training sequence and training the adaptive filter by using the training sequence; and filtering the voice signal by using a trained adaptive filter, wherein the howling component is suppressed. In the howling detection environment, the howling detection is carried out on the voice signal by utilizing the peak-to-average power ratio and the peak amplitude retention between frames, and the peak harmonic power ratio and the slope deviation of the amplitude spectrum between frames, so that the accuracy and the reliability of the howling detection are improved; meanwhile, the adaptive filtering method is adopted to suppress howling, so that the requirement on frequency estimation accuracy is reduced, and the voice quality is ensured.
Description
Technical Field
The invention belongs to the technical field of avionics, and particularly relates to a method, a device, equipment and a system for detecting and suppressing howling of an aircraft cockpit.
Background
On a civil aircraft, when the cabin speakers are turned on, the crew speaks through the microphones and the side tone (self-listening) is activated, the microphones, audio power amplifier, speakers form a typical amplification system. Because the cockpit is a narrow and tight space, the loudspeaker and the microphone are close to each other, and thus the phenomenon of howling is easy to generate.
Howling in the cockpit of a civil aircraft is a great hazard: firstly, the sharp and harsher whistle can cause discomfort to the crew, can submerge the sound of the crew voice system and the warning system, and can not make the pilot hear the important sounds, which seriously affects the flight safety; in addition, when the howling occurs, the audio equipment has extremely high output power, and the power amplifier and the loudspeaker of the airborne audio equipment are easy to burn for a long time, so that the service life of the equipment is shortened. Meanwhile, if the onboard audio system is disabled due to burning in the flight process, the flight safety is seriously affected.
Currently, the on-board audio systems of mainstream civil airliners are monopolized by foreign companies such as holmivir, kolin, and tylez in france in the united states, wherein key technologies for cockpit squeal suppression are also mastered by these companies to form technology blockages. At present, a strategic goal of the development of large domestic aircrafts in China is to improve the localization rate of airborne avionics and realize the autonomous controllability of key technologies, so that the formation of an airborne audio system with the autonomous research and development technology in China is urgently needed, and the problem of cockpit howling can be solved.
Disclosure of Invention
Aiming at the problems that the flight safety is affected, the service life of equipment is shortened and the like caused by the howling generated in the cockpit of the civil aircraft, the invention provides the method, the device, the equipment and the system for detecting and suppressing the howling of the cockpit of the aircraft, which make up the blank of the related technology in China, ensure the flight safety and improve the service life of the equipment.
The invention is realized by the following technical scheme:
an aircraft cockpit whistle detection and suppression method comprising:
the peak-to-average power ratio, the peak amplitude retention between frames, the peak harmonic power ratio and the slope deviation of the amplitude spectrum between frames are combined to detect howling on the voice signal;
after the occurrence of howling is detected, loading a training sequence and training the adaptive filter by using the training sequence;
and filtering the voice signal by using a trained adaptive filter, wherein the howling component is suppressed.
In the howling detection environment, the howling detection is carried out on the voice signal by utilizing the peak-to-average power ratio and the peak amplitude retention between frames, and the peak harmonic power ratio and the slope deviation of the amplitude spectrum between frames, so that the accuracy and the reliability of the howling detection are improved; meanwhile, the adaptive filtering method is adopted to suppress howling, so that the requirement on frequency estimation accuracy is reduced, and the voice quality is ensured.
As a preferred embodiment, the method for detecting howling of a voice signal according to the present invention specifically includes:
performing discrete Fourier transform on the voice signal frame and squaring a modulus value to obtain an energy spectrum of the voice signal frame;
calculating the peak-to-average power ratio of each frequency point and the peak amplitude retention between frames according to the energy spectrum, wherein the peak amplitude retention between frames is the number of frames in which the peak-to-average power ratio continuously exceeds a preset threshold;
comparing the amplitude maintenance degree of the peak value between frames with a preset threshold value, and screening out suspected howling frequency points;
carrying out howling judgment on all suspected howling frequency points smaller than N/2: calculating the peak harmonic power ratio of the suspected howling frequency point, if the peak harmonic power ratio exceeds a preset threshold, calculating the inter-frame amplitude spectrum slope deviation of the frequency point, and if the inter-frame amplitude spectrum slope deviation exceeds the preset threshold, judging the suspected howling frequency point as the howling frequency point; where N is the length of the voice signal frame.
As a preferred embodiment, when the peak-to-average power ratio of two adjacent frequency points exceeds a preset threshold value, the invention adopts the Paswal theorem to solve the problem of adjacent frequency ambiguity, and further improves the howling detection precision and reliability.
As a preferred embodiment, the method for howling determination on all the suspected howling frequency points smaller than N/2 specifically includes:
when calculating the peak harmonic power ratio of the suspected howling frequency point, if the suspected howling frequency point is smaller than N/4, 0.5, 1.5, 2, 3 and 4 subharmonics of the suspected howling frequency point are considered; if the suspected howling frequency point is greater than or equal to N/4 and less than N/2, 0.5, 1.5 and 2 harmonics of the suspected howling frequency point are considered;
calculating the ratio of the energy of the suspected howling frequency point to the energy of the harmonic wave considered correspondingly to obtain a peak harmonic wave power ratio, and if the peak harmonic wave power ratio exceeds a preset threshold value, keeping the judgment of the suspected howling frequency point;
calculating to obtain the inter-frame amplitude spectrum slope deviation of the reserved suspected howling frequency points according to the stored energy of the reserved suspected howling frequency points of the previous Q frames;
if the deviation of the inter-frame amplitude spectrum slope exceeds a preset threshold, the reserved suspected howling frequency point is judged to be the howling frequency point.
As a preferred embodiment, the method for loading a training sequence of the present invention specifically includes:
judging whether audio data to be played exist in an audio buffer;
if so, loading the audio data to be played as a training sequence, otherwise, generating a pseudo random sequence as the training sequence.
As a preferred embodiment, the adaptive filter of the present invention employs a finite impulse response filter of order L, with the filter coefficient length l+1, L being an even number.
As a preferred embodiment, the adaptive filter training process of the present invention comprises:
calculating an output of the filter;
calculating an error between an input and an output of the filter;
updating the filter coefficients based on the error.
In a second aspect, the invention provides a device for detecting and suppressing howling in an aircraft cockpit, which comprises a howling detection module and a howling suppression module;
the howling detection module carries out howling detection on the voice signal by adopting peak-to-average power ratio, inter-frame peak amplitude retention, peak harmonic power ratio and inter-frame amplitude spectrum slope deviation, and after howling is detected, a training sequence is loaded;
the howling suppression module comprises an adaptive filter, the adaptive filter is trained by using the training sequence, and the trained adaptive filter is adopted to filter the voice signal, so that the howling component is suppressed.
In a third aspect, the present invention proposes an on-board electronic device comprising a memory storing a computer program and a processor implementing the steps of the above-mentioned method of the present invention when said computer program is executed by said processor.
In a fourth aspect, the present invention provides an aircraft cockpit audio system, including the above-mentioned on-board electronic device of the present invention;
the airborne electronic equipment is used for detecting and suppressing howling of the microphone signal and outputting the microphone signal to the loudspeaker for playing;
the voice signals played by the loudspeaker are collected by the microphone after passing through the sound field of the cockpit.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. in the howling detection link, according to the characteristics of the sound link in the cockpit during flight, namely that the noise at the bottom is larger and the dynamic range is larger, the crew can adjust the speaking loudness at any time according to the noise at the bottom in the cockpit, so that the dynamic range of the microphone input voice signal is enlarged, and the invention adopts the peak-to-average power ratio (PAPR) and the inter-frame peak amplitude maintenance (IPMP) as indexes for preliminary judgment of the howling; simultaneously combines the characteristics of the howling signal: the harmonic component of the howling signal is less, even if the howling signal makes the public address system saturated and peak clipping occurs, the harmonic energy is much lower than the fundamental wave energy, and the howling signal is longer than the duration of the voice signal, therefore, the invention utilizes Peak Harmonic Power Ratio (PHPR) and inter-frame amplitude spectrum slope deviation (IMSD) to further reject the voice signal which is falsely detected as howling, and ensures the precision and reliability of the howling detection;
2. in the howling suppression link, a method based on detection-suppression generates a corresponding Infinite Impulse Response (IIR) according to the howling frequency to suppress howling frequency points, the existing method for detecting the howling frequency by using a wave trap can influence the suppression effect of the wave trap on the howling, and even if the Q value of a filter is reduced to reduce the requirement on the frequency estimation precision, more voice signals can be filtered to reduce the tone quality, so that the invention adopts an adaptive filtering method to suppress the howling.
3. Compared with the conventional method that a pseudo-random sequence is generated through a trainer, the self-adaptive filtering method provided by the invention simultaneously has the functional characteristics of an airborne audio signal set sum mixing device according to airborne audio processing equipment, and when audio data to be played transmitted by other airborne equipment exist in a digital audio buffer, the data are used as training sequences, so that noise introduced into an audio system in the training process can be reduced, otherwise, the pseudo-random sequence is generated according to the conventional method, and the performance of the audio system is ensured.
Drawings
The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention. In the drawings:
FIG. 1 is a schematic overall flow chart of a method according to an embodiment of the invention.
Fig. 2 is a schematic diagram of a howling detection flow according to an embodiment of the present invention.
FIG. 3 is a plot of the continuous energy spectrum versus the discrete energy spectrum of the DFT for a finite time sinusoidal signal in accordance with an embodiment of the present invention.
Fig. 4 is a schematic block diagram of an apparatus according to an embodiment of the present invention.
Detailed Description
For the purpose of making apparent the objects, technical solutions and advantages of the present invention, the present invention will be further described in detail with reference to the following examples and the accompanying drawings, wherein the exemplary embodiments of the present invention and the descriptions thereof are for illustrating the present invention only and are not to be construed as limiting the present invention.
Examples:
in the existing howling detection and suppression technology, the problem of adjacent frequency blurring exists in a detection link, so that the estimation of howling frequency points is inaccurate due to insufficient frequency spectrum resolution; in addition, the effect of the existing howling suppression technology in the cockpit environment of the civil aircraft is not ideal, for example, measures such as redesigning the position of a cockpit loudspeaker, properly reducing microphone sensitivity and loudspeaker volume and the like can improve howling but cannot fundamentally realize suppression, and meanwhile, the side sound volume is reduced to influence the use experience of pilots; the phase shifting method has limited capability of suppressing howling in a closed and narrow cockpit; the trap suppression method has high accuracy requirements for howling detection and howling frequency estimation, and high accuracy frequency estimation has high performance requirements for the computational processing unit. Based on this, this embodiment provides a method for detecting and suppressing howling in an aircraft cockpit
As shown in fig. 1, the method provided in this embodiment includes the following steps:
step 3, the voice signal frames are passed through a trained adaptive filter, and the howling component is suppressed.
In an alternative embodiment, when howling is detected, firstly, discrete fourier transform is performed on the voice signal frame and the square of the modulus value is obtained to obtain the energy spectrum of the voice signal frame; then, calculating Peak-to-Average Power Ratio (PAPR) and inter-frame Peak amplitude retention (Interframe Peak Magnitude Persistence, IPMP, i.e. the number of frames in which the PAPR continuously exceeds a preset threshold) of each frequency point according to the energy spectrum; comparing the inter-frame peak amplitude maintenance degree IPMP with a preset threshold value, and screening out suspected howling frequency points; for the suspected howling frequency points, calculating Peak-to-Harmonic Power Ratio (PHPR) of the suspected howling frequency points; if the peak harmonic power ratio PHRP exceeds a preset threshold, further calculating the inter-frame amplitude spectrum slope deviation (Interframe Magnitude Slope Deviation, IMSD) of the frequency point; if the deviation of the inter-frame amplitude spectrum slope exceeds a preset threshold, the suspected howling frequency point is judged to be the howling frequency point. After the above judgment is completed for all the frequency bands smaller than N/2, if howling frequency points exist, the subsequent operation of training sequence loading and adaptive filter training is started to be executed.
In the complex low-noise environment, compared with the existing technology for detecting howling by using PNPR (when the howling frequency is located between two DFT frequency points, the amplitudes of the two DFT frequency points are similar, surrounding DFT frequency points are also located at the center position of a sinc frequency spectrum sidelobe, and the amplitudes are not low enough, so that the detection of the howling by PNPR is easy to cause), the embodiment realizes the howling detection by using PAPR, and can better solve the problem of adjacent frequency ambiguity caused by insufficient DFT sampling points.
As shown in fig. 2, the howling detection process specifically includes the following steps:
for a length of n=2 M T-th frame microphone input signal x [ n, t ] (M is a positive integer)]N=0, 1,..n, N performs an N-point Discrete Fourier Transform (DFT)
Then square the modulus to obtain the energy spectrum Y [ k, t ] of the frame signal]＝|X[k，t]| 2 . Among them, DFT may be implemented by a Fast Fourier Transform (FFT) algorithm.
Thereafter, for each frequency bin k (k=0, 1,2 … N/2-1), it is determined whether howling occurs or not as follows:
(1) For the frequency point k, the PAPR is calculated according to the following formula:
if PAPR exceeds the thresholdThreshold T set in advance PAPR PAPR [ k, t ]]>T PAPR Then IPMP [ k, t]＝IPMP[k，t-1]+1, and recording the energy Y [ k ] of the frequency point]The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, IPMP [ k, t]And (5) returning to zero.
The following phenomenon occurs when calculating the PAPR of each DFT frequency point: the PAPR with two adjacent frequency points k and k+1 exceeds the threshold at the same time, and this phenomenon occurs because the howling frequency is generally located between two frequencies represented by some two adjacent frequency points, and does not coincide with the frequency represented by some frequency point. As shown in FIG. 3, the duration is N/f s Has a frequency f 0 Can characterize the howling frequency point f 0 Is a sinc function that shifts the zero frequency point around, with a main lobe width f) s N. Therefore, the k and k+1 frequency points of the DFT discrete spectrum sample the main lobe of the continuous spectrum, and the other frequency points sample the side lobes. While the main lobe occupies almost all the energy of the signal, so that the energy at the howling frequency can be approximated as Y k according to the Paswal theorem]+Y[k+1]。
Therefore, the present embodiment solves the problem of adjacent frequency ambiguity based on the paswal theorem, and when the PAPR of two adjacent frequency points k and k+1 exceeds the threshold simultaneously, one of the frequency points is reserved according to the following rule:
a) If Y [ k, t ] > Y [ k+1, t ], then keep k, record Y [ k, t ] = Y [ k, t ] + Y [ k+1, t ], IPMP [ k, t ] = max { IPMP [ k, t-1], IPMP [ k+1, t-1] } +1, IPMP [ k+1, t ] = 0;
b) Otherwise, k+1 is reserved, Y [ k+1, t ] =Y [ k, t ] +Y [ k+1, t ], IPMP [ k+1, t ] =max { IPMP [ k, t-1], IPMP [ k+1, t-1] } +1, IPMP [ k, t ] =0.
(2) If the IPMP of the frequency point k exceeds the preset threshold T IPMP I.e. IPMP [ k, t]＞T IPMP And marking the frequency point as a suspected howling frequency point, and carrying out subsequent howling judgment.
(3) PHPR criterion howling decision. When PHPR of the suspected howling frequency point is calculated, if k is less than N/4, 0.5, 1.5, 2, 3 and 4 harmonics of the frequency point k are considered; if N/4.ltoreq.k < N/2, 0.5, 1.5 and 2 harmonics of the frequency bin k are considered. The energy of the frequency point k is equal to the energy of the frequency point kThe ratio of the energy of the harmonic wave exceeds a preset threshold T PHPR The judgment that the frequency point k is the howling frequency point is reserved, the subsequent howling judgment is carried out, and the judgment process of PHPR can be described as the following formula:
if k > N/2, the PHPR decision is skipped and a subsequent howling decision is made.
(4) According to the stored energy of the frequency point k of the previous Q frame, Y [ k, t-Q ], Y [ k, t-Q+1, &..Yk, t-1, the IMSD of the frequency point k is calculated, namely:
if the preset threshold is exceeded, i.e. IMSD [ k, t]＞T IMSD And judging the frequency point k as a howling frequency point.
In an alternative implementation manner, the suppression of the howling frequency point is realized by generating a corresponding Infinite Impulse Response (IIR) according to the howling frequency, and the accuracy requirements of a wave trap method on howling detection and howling frequency estimation are higher; based on this, in this embodiment, after howling detection is completed and howling is detected, an adaptive filtering method is used to perform howling suppression, specifically, according to whether audio data from other aircraft avionics systems exist in the buffer memory, the adaptive filter is trained by selecting the loading audio data or the pseudo-random sequence as a training sequence, and then, when the microphone signal passes through the trained adaptive filter, the howling component can be suppressed.
The adaptive filter of the embodiment adopts a Finite Impulse Response (FIR) filter, and compared with an IIR filter, the adaptive filter has no feedback loop, does not generate parasitic oscillation, and has strong stability; the FIR has a linear phase, the distortion to the sound is small, and the degree of dependence on the spectrum analysis accuracy is not high; compared with a wave trap method, the spectrum analysis of the adaptive filtering method of the embodiment only needs to detect whether howling exists or not, does not need to estimate the howling frequency, and can adaptively and autonomously generate a filter suitable for the current sound field environment, so that the method is not influenced by the accuracy of the estimation of the howling frequency, and the performance of an audio system is improved.
The principle of training the adaptive filter is: and a section of known training sequence is passed through the whole system, and the coefficients of the filter are continuously adjusted according to the difference between the system output and the original training sequence, so that the difference is smaller and smaller. Therefore, any sequence known in advance can be used for training. The airborne audio processing equipment has the function of summation and mixing of airborne audio signal sets, and when audio data to be played transmitted by other airborne equipment exist in the digital audio buffer, the audio data are used as training sequences, so that noise introduced to an audio system in the training process can be reduced.
The loading method of the training sequence specifically comprises the following steps:
(1) Judging whether audio data to be played exist in an audio buffer memory of the airborne audio processing equipment;
(2) If yes, loading audio data to be played as a training sequence; otherwise, generating a pseudo-random sequence as a training sequence.
In this embodiment, the LMS algorithm is used to perform adaptive filter training, where the training starts at time t=1, and the training ends until time T. The t-th value of the training sequence is s (t), and the length is L of the column vector
And->
Respectively representing the input signal and the filter coefficients stored in the shift register of the filter at time t. The step of updating and calculating the filter at the t-th moment is as follows:
(1) Calculating the output of the filter:
(2) Calculating errors
e(t)＝s(t)-y(t)
(3) For filter coefficients
Updating:
where μ represents the step size of the filter coefficient update, and μ=0.001 is taken in this embodiment.
The embodiment also provides a device for detecting and suppressing the howling of the cockpit of the aircraft, and particularly as shown in fig. 4, the system comprises a howling detection module and a howling suppression module;
the howling detection module is used for detecting howling of a voice signal frame (namely a microphone signal frame), and after the occurrence of howling is detected, a training sequence is loaded;
the howling suppression module comprises a delay unit and an adaptive filter; the delay unit delays the training sequence and inputs the training sequence into the adaptive filter to train the adaptive filter. The training algorithm is based on comparing the output and input of an adaptive filter, which is a Finite Impulse Response (FIR) filter with an order of L, to adjust the coefficients of the adaptive filter, the filter coefficients having a length of l+1, typically L being an even number; the delay is set because the output of the FIR filter has a lag of L/2+1 points relative to the input, and the adaptive filter coefficients are initialized (n takes an integer from 1 to l+1):
the trained adaptive filter filters the voice signal frames and the howling component is suppressed.
The embodiment also provides an onboard electronic device for executing the method of the embodiment.
The embodiment also provides an aircraft cockpit audio system, which comprises a microphone, a loudspeaker and the airborne electronic equipment, wherein the airborne electronic equipment is used for carrying out howling detection and suppression on microphone signals and then outputting the microphone signals to the loudspeaker for playing, and then the microphone signals are collected after passing through the sound field of the cockpit, so that the whole process of acoustic feedback is completed.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application 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, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (10)
1. An aircraft cockpit whistle detection and suppression method, comprising:
the peak-to-average power ratio, the peak amplitude retention between frames, the peak harmonic power ratio and the slope deviation of the amplitude spectrum between frames are combined to detect howling on the voice signal;
after the occurrence of howling is detected, loading a training sequence and training the adaptive filter by using the training sequence;
and filtering the voice signal by using a trained adaptive filter, wherein the howling component is suppressed.
2. The method for detecting and suppressing howling in an aircraft cockpit according to claim 1, wherein the howling detection is performed on the speech signal, specifically comprising:
performing discrete Fourier transform on the voice signal frame and squaring a modulus value to obtain an energy spectrum of the voice signal frame;
calculating the peak-to-average power ratio of each frequency point and the peak amplitude retention between frames according to the energy spectrum, wherein the peak amplitude retention between frames is the number of frames in which the peak-to-average power ratio continuously exceeds a preset threshold;
comparing the amplitude maintenance degree of the peak value between frames with a preset threshold value, and screening out suspected howling frequency points;
carrying out howling judgment on all suspected howling frequency points smaller than N/2: calculating the peak harmonic power ratio of the suspected howling frequency point, if the peak harmonic power ratio exceeds a preset threshold, calculating the inter-frame amplitude spectrum slope deviation of the frequency point, and if the inter-frame amplitude spectrum slope deviation exceeds the preset threshold, judging the suspected howling frequency point as the howling frequency point; where N is the length of the voice signal frame.
3. The method for detecting and suppressing howling in an aircraft cockpit according to claim 2, wherein when the peak-to-average power ratio of two adjacent frequency points exceeds a preset threshold value at the same time, the problem of adjacent frequency ambiguity is solved by using the paswal theorem.
4. The method for detecting and suppressing howling in an aircraft cockpit according to claim 2, wherein the howling determination is performed on all the suspected howling frequency points smaller than N/2, specifically comprising:
when calculating the peak harmonic power ratio of the suspected howling frequency point, if the suspected howling frequency point is smaller than N/4, 0.5, 1.5, 2, 3 and 4 subharmonics of the suspected howling frequency point are considered; if the suspected howling frequency point is greater than or equal to N/4 and less than N/2, 0.5, 1.5 and 2 harmonics of the suspected howling frequency point are considered;
calculating the ratio of the energy of the suspected howling frequency point to the energy of the harmonic wave considered correspondingly to obtain a peak harmonic wave power ratio, and if the peak harmonic wave power ratio exceeds a preset threshold value, keeping the judgment of the suspected howling frequency point;
calculating to obtain the inter-frame amplitude spectrum slope deviation of the reserved suspected howling frequency points according to the stored energy of the reserved suspected howling frequency points of the previous Q frames;
if the deviation of the inter-frame amplitude spectrum slope exceeds a preset threshold, the reserved suspected howling frequency point is judged to be the howling frequency point.
5. An aircraft cockpit whistle detection and suppression method according to any of claims 1-4, characterized in that the method for loading the training sequence specifically comprises:
judging whether audio data to be played exist in an audio buffer;
if so, loading the audio data to be played as a training sequence, otherwise, generating a pseudo random sequence as the training sequence.
6. The method for detecting and suppressing howling in an aircraft cockpit according to claim 5, wherein said adaptive filter uses a finite impulse response filter of order L, the filter coefficient length being l+1, L being an even number.
7. The aircraft cockpit howling detection and suppression method of claim 6 wherein said adaptive filter training process comprises:
calculating an output of the filter;
calculating an error between an input and an output of the filter;
updating the filter coefficients based on the error.
8. The device for detecting and suppressing the howling of the aircraft cockpit is characterized by comprising a howling detection module and a howling suppression module;
the howling detection module carries out howling detection on the voice signal by adopting peak-to-average power ratio, inter-frame peak amplitude retention, peak harmonic power ratio and inter-frame amplitude spectrum slope deviation, and after howling is detected, a training sequence is loaded;
the howling suppression module comprises an adaptive filter, the adaptive filter is trained by using the training sequence, and the trained adaptive filter is adopted to filter the voice signal, so that the howling component is suppressed.
9. An on-board electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-7 when the computer program is executed.
10. An aircraft cockpit audio system comprising the on-board electronics of claim 9;
the airborne electronic equipment is used for detecting and suppressing howling of the microphone signal and outputting the microphone signal to the loudspeaker for playing;
the voice signals played by the loudspeaker are collected by the microphone after passing through the sound field of the cockpit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273614.6A CN116312597A (en) | 2023-03-20 | 2023-03-20 | Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273614.6A CN116312597A (en) | 2023-03-20 | 2023-03-20 | Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116312597A true CN116312597A (en) | 2023-06-23 |
Family
ID=86786712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310273614.6A Pending CN116312597A (en) | 2023-03-20 | 2023-03-20 | Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116312597A (en) |
-
2023
- 2023-03-20 CN CN202310273614.6A patent/CN116312597A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11017799B2 (en) | Method for processing voice in interior environment of vehicle and electronic device using noise data based on input signal to noise ratio | |
| US8170879B2 (en) | Periodic signal enhancement system | |
| KR101798120B1 (en) | Apparatus and method for improving the perceived quality of sound reproduction by combining active noise cancellation and perceptual noise compensation | |
| US7610196B2 (en) | Periodic signal enhancement system | |
| EP2364037B1 (en) | Adaptive notch filter with variable bandwidth, and method and apparatus for canceling howling by using the adaptive notch filter with variable bandwidth | |
| US9414157B2 (en) | Method and device for reducing voice reverberation based on double microphones | |
| JP2012212161A (en) | Active noise control system using audio signal compensation | |
| US9031248B2 (en) | Vehicle engine sound extraction and reproduction | |
| US8306821B2 (en) | Sub-band periodic signal enhancement system | |
| US11238882B2 (en) | Dry sound and ambient sound separation | |
| CN112037816A (en) | Voice signal frequency domain frequency correction, howling detection and suppression method and device | |
| CA3055447A1 (en) | System and method for audio noise reduction | |
| US11380312B1 (en) | Residual echo suppression for keyword detection | |
| CN103929704B (en) | The method and system that a kind of adaptive acoustic feedback based on transform domain is eliminated | |
| KR20190047976A (en) | Method of Noise Decresing Using Noise Modelling and Lookup | |
| US10937418B1 (en) | Echo cancellation by acoustic playback estimation | |
| JP2006243644A (en) | Method for reducing noise, device, program, and recording medium | |
| JP5762479B2 (en) | Voice switch device, voice switch method, and program thereof | |
| US9959852B2 (en) | Vehicle engine sound extraction | |
| US11386911B1 (en) | Dereverberation and noise reduction | |
| KR101587844B1 (en) | Microphone signal compensation apparatus and method of the same | |
| JP4690243B2 (en) | Digital filter, periodic noise reduction device, and noise reduction device | |
| JPH11265199A (en) | Voice transmitter | |
| JP5466581B2 (en) | Echo canceling method, echo canceling apparatus, and echo canceling program | |
| CN116312597A (en) | Method, device, equipment and system for detecting and suppressing howling of aircraft cockpit |
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 |