Communication quality intelligent tuning method and system applied to dual-mode interphone

Abstract

The invention relates to the field of interphone communication noise adjustment, and discloses a communication quality intelligent tuning method and system applied to a dual-mode interphone, and the method comprises the following steps: carrying out the noise reduction and segmentation of an input audio signal in the dual-mode interphone, the method comprises the following steps: obtaining audio signals of human voice and environment voice, encoding the audio signals into digital signals, transmitting the digital signals to other dual-mode interphones for decoding, carrying out noise analysis on the decoded audio signals, and finally carrying out noise tracing and communication quality optimization on the audio signals based on a noise analysis result. According to the invention, noise processing can be carried out on the audio signal, so that the purposes of improving the communication efficiency between the dual-mode interphones, optimizing the user experience and adapting to a complex environment are achieved, and the communication quality is optimized at the same time.

Description

Technical Field

[0001] The invention relates to the field of intercom communication noise regulation, and in particular to a communication quality intelligent tuning method and system applied to a dual-mode intercom. Background Art

[0002] Dual-mode walkie-talkies refer to devices that are compatible with two or more communication modes, such as analog and digital modes, public network and private network modes. In analog mode, dual-mode walkie-talkies use frequency modulation (FM) technology to send and receive signals. In this mode, information is transmitted in the form of analog signals of radio waves, which has the advantages of long transmission distance and strong signal penetration; in digital mode, dual-mode walkie-talkies use digital encoding and decoding technology to transmit information in digital form. In this mode, communication has the characteristics of strong anti-interference ability and high communication quality.

[0003] In a noisy environment, background noise and echoes can interfere with communication quality and make communication difficult. Through intelligent tuning, dual-mode walkie-talkies can use noise reduction algorithms and echo cancellation technology to effectively reduce background noise and eliminate echoes, thereby improving the clarity and audibility of communication. Among them, the noise reduction algorithm is particularly important because dual-mode walkie-talkies often need to be used in changing communication environments, such as high-rise buildings in cities, deep mountains and dense forests. Intelligent tuning of the noise reduction algorithm can ensure that the walkie-talkie can still maintain stable communication quality in these complex environments. At the same time, as spectrum resources become increasingly tight, dual-mode walkie-talkies can use spectrum resources more effectively through intelligent tuning. This helps reduce spectrum waste and improve the overall efficiency and capacity of the communication system. Summary of the invention

[0004] The present invention overcomes the deficiencies of the prior art and provides a communication quality intelligent tuning method and system for a dual-mode intercom.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A first aspect of the present invention provides a communication quality intelligent tuning method applied to a dual-mode intercom, comprising the following steps: Perform feature processing and feature analysis on the audio signal input into the dual-mode intercom to achieve noise reduction processing of the audio signal and obtain the target noise reduction audio signal; Performing audio feature analysis and audio segmentation on the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively; Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and perform noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal; When there is noise in the received audio signal, it is necessary to combine the receiving dual-mode walkie-talkie and the surrounding environment to trace the noise source and optimize the communication quality of the received audio signal.

[0006] Further, in a preferred embodiment of the present invention, the feature processing and feature analysis of the audio signal input into the dual-mode intercom are performed to realize the noise reduction processing of the audio signal to obtain the target noise reduction audio signal, specifically: A dual-mode intercom that needs to be tuned for communication quality is obtained and calibrated as a target dual-mode intercom, wherein the target dual-mode intercom can collect audio and convert it into an audio signal, and perform encoding, decoding and feature analysis on the audio signal; Running the target dual-mode intercom to enable the target dual-mode intercom to collect audio signals in real time, and calibrating the audio signals as target audio signals, wherein the target audio signals include environmental audio signals and human voice audio signals; In the target dual-mode intercom, a target audio signal is subjected to DC component preprocessing and audio amplification to obtain a preprocessed target audio signal, the preprocessed target audio signal is subjected to Hanning window processing, and the preprocessed target audio signal is subjected to short-time Fourier transform after the Hanning window processing to generate a spectrum diagram of the preprocessed target audio signal, which is calibrated as a target spectrum diagram; In the target spectrum schematic diagram, quantile modeling analysis is performed on different frequency points on the target spectrum schematic diagram by using the quantile noise statistics method, and a quantile threshold is preset, and the frequency points whose quantiles are maintained within the quantile threshold are marked as noise signal frequency points, and the noise signal frequency points are marked and the target spectrum schematic diagram is updated to obtain an updated spectrum schematic diagram, wherein the updated spectrum schematic diagram records the spectrum characteristics of the noise signal; According to the spectral characteristics of the noise signal, the corresponding spectral flatness is calculated at the corresponding noise signal frequency point, wherein the spectral flatness is the difference between the noise spectral characteristics and the non-noise spectral characteristics at the noise signal frequency point, and based on the spectral flatness of different noise signal frequency points, the noise suppression coefficient of different noise signal frequency points is calculated by weighted average method; The noise suppression coefficients of different noise signal frequencies are imported into the updated spectrum diagram and spectrally multiplied with the corresponding noise signal frequencies to obtain the updated spectrum diagram after noise reduction. The updated spectrum diagram after noise reduction is subjected to an inverse short-time Fourier transform to obtain the target audio signal after noise reduction, which is calibrated as the target noise reduction audio signal.

[0007] Furthermore, in a preferred embodiment of the present invention, the target noise reduction audio signal is subjected to audio feature analysis and audio segmentation, and the segmented audio signals are encoded into digital signals, specifically: Introducing a CNN neural network, wherein the CNN neural network includes a convolutional layer, a pooling layer and a fully connected layer; The sampling rate of the target noise reduction audio signal is adjusted to a sampling rate that the CNN neural network can perform sampling analysis on, thereby obtaining a sampled and adjusted noise reduction audio signal; Performing Mel spectrum conversion on the sampled and regulated noise-reduced audio signal to convert it into a sampled and regulated noise-reduced audio signal in the frequency domain state, importing the sampled and regulated noise-reduced audio signal in the frequency domain state into the convolution layer of the CNN neural network for convolution processing, and reducing the dimension of the sampled and regulated noise-reduced audio signal in the frequency domain state after the convolution processing through the pooling layer to realize local feature extraction of the sampled and regulated noise-reduced audio signal in the frequency domain state; Through the fully connected layer, the local features of the extracted sampled and regulated noise reduction audio signal are mapped to the human voice output category and the ambient sound output category to obtain the human voice audio signal and the ambient sound audio signal, which are calibrated as the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal; Determine the amplitude range and frequency range of the digital signal that the target dual-mode intercom can meet when demodulating the digital signal, calibrate them as the standard amplitude range and standard frequency range, and in the target dual-mode intercom, respectively perform amplitude modulation and frequency modulation on the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal, so that the amplitude and frequency of the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal meet the standard amplitude range and standard frequency range respectively, and obtain the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

[0008] Further, in a preferred embodiment of the present invention, the control target dual-mode intercom transmits the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and performs noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, specifically: Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, and calibrate the dual-mode intercom that receives the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal as the receiving dual-mode intercom; In the receiving dual-mode intercom, a target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal are received, and the received target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal are collectively referred to as received digital signals; In the receiving dual-mode intercom, adaptive differential pulse code modulation is performed on the received digital signal to decode the received digital signal into an audio signal to obtain a received audio signal, wherein the received audio signal includes a received target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal, which are calibrated as a received target noise reduction human voice audio signal and a received target noise reduction environmental sound audio signal; In the receiving dual-mode intercom, the received audio signal is converted into a spectrum to generate a received audio signal spectrum, a big data network is obtained, a standard frequency component range of the received audio signal is retrieved based on the big data network, and the received audio signal spectrum is analyzed. If there are frequency components that are not maintained within the standard frequency component range or there are clutters on the received audio signal spectrum, it is determined that the received audio signal has noise; If there is no frequency component that does not remain within the standard frequency component range on the received audio signal spectrum and no clutter exists, it is determined that there is no noise in the received audio signal; If there is no noise in the received audio signal, the receiving dual-mode intercom will give priority to outputting the receiving target noise reduction human voice audio signal, and determine whether the receiving target noise reduction ambient sound audio signal can be output in the receiving dual-mode intercom; If so, the received target noise reduction human voice audio signal and the received target noise reduction ambient sound audio signal are output simultaneously; if not, only the received target noise reduction human voice audio signal is output.

[0009] Furthermore, in a preferred embodiment of the present invention, when there is noise in the received audio signal, it is necessary to combine the receiving dual-mode intercom and the surrounding environment to trace the noise source of the received audio signal and optimize the communication quality, specifically: Obtain the communication range between the receiving dual-mode intercom and the target dual-mode intercom, and mark it as the target communication range; An environmental monitoring platform is determined in the big data network, and based on the environmental monitoring platform, environmental parameters in the target communication range are determined, calibrated as target communication range environmental parameters, and the real-time minimum communicable distance of the target noise-reduced human voice digital signal and the target noise-reduced environmental sound digital signal in the target communication range is measured, and the standard maximum communicable distance is determined at the same time; If the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal in the target communication range is greater than the standard maximum communicable distance, it is determined that the received audio signal has noise, which is an abnormality in the target communication range; Perform distance control on the target dual-mode intercom and the receiving dual-mode intercom, so that the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal is not greater than the standard maximum communicable distance, and continue to analyze the received audio signal spectrum; If there are still frequency components that are not maintained within the standard frequency component range on the received audio signal spectrum, or there are clutter, then the appropriate communication environment parameters of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal are retrieved in the big data network, and the scheme for adjusting the target communication range environment parameters so that the target communication range environment parameters are maintained equal to the appropriate communication environment parameters is retrieved and output; If there are still frequency components that are not maintained within the standard frequency component range or noise on the received audio signal spectrum, the communication frequency band optimization processing is performed between the receiving dual-mode walkie-talkie and the target dual-mode walkie-talkie to achieve communication quality optimization.

[0010] Further, in a preferred embodiment of the present invention, the communication frequency band optimization processing is performed between the receiving dual-mode intercom and the target dual-mode intercom to achieve communication quality optimization, specifically: Determine the standard communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom, and check the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom to determine whether the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are greater than the standard communication frequency band; If not, the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are adjusted to make the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom equal to the standard communication frequency band; If the real-time communication frequency band of the receiving dual-mode intercom and the target dual-mode intercom is equal to the standard communication frequency band, but there are still frequency components that are not maintained within the standard frequency component range or noise in the received audio signal spectrum, the digital error correction technology is combined on the receiving dual-mode intercom to filter the noise of the received audio signal, and the signal of the received audio signal after the noise filtering is reconstructed, and finally the received audio signal after the signal reconstruction and noise filtering is output; The signal reconstruction is to adjust the frequency of the received audio signal after noise filtering so that the frequency of the received audio signal after noise filtering is equal to the frequency of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

[0011] The second aspect of the present invention further provides a communication quality intelligent tuning system applied to a dual-mode intercom, the communication quality intelligent tuning system comprising a memory and a processor, the memory storing a communication quality intelligent tuning method, and when the communication quality intelligent tuning method is executed by the processor, the following steps are implemented: Perform feature processing and feature analysis on the audio signal input into the dual-mode intercom to achieve noise reduction processing of the audio signal and obtain the target noise reduction audio signal; Performing audio feature analysis and audio segmentation on the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively; Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and perform noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal; When there is noise in the received audio signal, it is necessary to combine the receiving dual-mode walkie-talkie and the surrounding environment to trace the noise source and optimize the communication quality of the received audio signal.

[0012] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: in the dual-mode intercom, the input audio signal is subjected to noise reduction and segmentation processing, and the audio signals of human voice and environmental sound are obtained and then encoded into digital signals and transmitted to other dual-mode intercoms for decoding, and the decoded audio signals are subjected to noise analysis, and finally the audio signals are subjected to noise tracing and communication quality optimization based on the noise analysis results. The present invention can achieve the purpose of improving communication efficiency, optimizing user experience and adapting to complex environments between dual-mode intercoms by performing noise processing on audio signals, while optimizing communication quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0014] Figure 1 A flow chart of a communication quality intelligent tuning method applied to a dual-mode intercom is shown; Figure 2 A flow chart of a method for tracing the source of noise on a received audio signal and optimizing communication quality is shown; Figure 3 A program view of a communication quality intelligent tuning system applied to a dual-mode intercom is shown. DETAILED DESCRIPTION

[0015] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0016] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0017] Figure 1 A flow chart of a method for intelligently optimizing communication quality of a dual-mode intercom is shown, comprising the following steps: S102: performing feature processing and feature analysis on the audio signal input into the dual-mode intercom to implement noise reduction processing on the audio signal and obtain a target noise reduction audio signal; S104: performing audio feature analysis and audio segmentation on the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively; S106: Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and perform noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal; S108: When there is noise in the received audio signal, it is necessary to perform noise tracing and communication quality optimization on the received audio signal in combination with the receiving dual-mode intercom and the surrounding environment.

[0018] Further, in a preferred embodiment of the present invention, the feature processing and feature analysis of the audio signal input into the dual-mode intercom are performed to realize the noise reduction processing of the audio signal to obtain the target noise reduction audio signal, specifically: A dual-mode intercom that needs to be tuned for communication quality is obtained and calibrated as a target dual-mode intercom, wherein the target dual-mode intercom can collect audio and convert it into an audio signal, and perform encoding, decoding and feature analysis on the audio signal; Running the target dual-mode intercom to enable the target dual-mode intercom to collect audio signals in real time, and calibrating the audio signals as target audio signals, wherein the target audio signals include environmental audio signals and human voice audio signals; In the target dual-mode intercom, a target audio signal is subjected to DC component preprocessing and audio amplification to obtain a preprocessed target audio signal, the preprocessed target audio signal is subjected to Hanning window processing, and the preprocessed target audio signal is subjected to short-time Fourier transform after the Hanning window processing to generate a spectrum diagram of the preprocessed target audio signal, which is calibrated as a target spectrum diagram; In the target spectrum schematic diagram, quantile modeling analysis is performed on different frequency points on the target spectrum schematic diagram by using the quantile noise statistics method, and a quantile threshold is preset, and the frequency points whose quantiles are maintained within the quantile threshold are marked as noise signal frequency points, and the noise signal frequency points are marked and the target spectrum schematic diagram is updated to obtain an updated spectrum schematic diagram, wherein the updated spectrum schematic diagram records the spectrum characteristics of the noise signal; According to the spectral characteristics of the noise signal, the corresponding spectral flatness is calculated at the corresponding noise signal frequency point, wherein the spectral flatness is the difference between the noise spectral characteristics and the non-noise spectral characteristics at the noise signal frequency point, and based on the spectral flatness of different noise signal frequency points, the noise suppression coefficient of different noise signal frequency points is calculated by weighted average method; The noise suppression coefficients of different noise signal frequencies are imported into the updated spectrum diagram and spectrally multiplied with the corresponding noise signal frequencies to obtain the updated spectrum diagram after noise reduction. The updated spectrum diagram after noise reduction is subjected to an inverse short-time Fourier transform to obtain the target audio signal after noise reduction, which is calibrated as the target noise reduction audio signal.

[0019] It should be noted that the function of the dual-mode walkie-talkie is to perform noise reduction and codec processing on the collected audio signal. In order to transmit the input audio signal losslessly, it is necessary to perform advanced noise reduction processing on the collected audio signal, and then encode it into a digital signal. The audio signal input to the walkie-talkie includes human voice and environmental sound, and the main audio transmitted is human voice. The statistical model is mainly used to reduce the noise of the audio signal. The noise and voice components corresponding to each frequency point in the audio spectrum are estimated by statistical methods to achieve noise reduction. First, the audio signal is collected and the DC component of the audio signal is processed to improve the accuracy of subsequent processing. Then, the pre-processed noisy audio signal is subjected to Hanning window processing to improve the possible missing parts of the audio signal, and a fast short-time Fourier transform is performed to obtain the spectrum representation of the noisy audio signal. Then, a statistical method is used for modeling, namely the quantile noise statistics method, which usually involves analyzing the spectral characteristics of the signal during a silent or noise-dominated time period. If the quantile does not match the quantile threshold, there is a noise signal frequency point. In the target spectrum diagram, the noise signal frequency point indicates that the point in the target spectrum diagram is a noise point, and then the updated spectrum diagram is obtained. According to the spectral characteristics of the noisy signal, the probability of speech existence is estimated, which involves calculating the spectrum flatness. Finally, the noise suppression coefficients of different noise signal frequency points are calculated by weighted averaging based on the updated spectrum diagram and the probability of speech existence, that is, combined with the spectrum flatness. These coefficients are used to adjust the spectrum of the noisy signal in subsequent steps to suppress noise. The noise suppression coefficient of each frequency point is multiplied by the spectrum of the noisy signal to obtain the spectrum representation after noise reduction. The noise reduction spectrum is subjected to an inverse short-time Fourier transform to obtain the noise-reduced audio signal.

[0020] Furthermore, in a preferred embodiment of the present invention, the target noise reduction audio signal is subjected to audio feature analysis and audio segmentation, and the segmented audio signals are encoded into digital signals, specifically: Introducing a CNN neural network, wherein the CNN neural network includes a convolutional layer, a pooling layer and a fully connected layer; The sampling rate of the target noise reduction audio signal is adjusted to a sampling rate that the CNN neural network can perform sampling analysis on, thereby obtaining a sampled and adjusted noise reduction audio signal; Performing Mel spectrum conversion on the sampled and regulated noise-reduced audio signal to convert it into a sampled and regulated noise-reduced audio signal in the frequency domain state, importing the sampled and regulated noise-reduced audio signal in the frequency domain state into the convolution layer of the CNN neural network for convolution processing, and reducing the dimension of the sampled and regulated noise-reduced audio signal in the frequency domain state after the convolution processing through the pooling layer to realize local feature extraction of the sampled and regulated noise-reduced audio signal in the frequency domain state; Through the fully connected layer, the local features of the extracted sampled and regulated noise reduction audio signal are mapped to the human voice output category and the ambient sound output category to obtain the human voice audio signal and the ambient sound audio signal, which are calibrated as the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal; Determine the amplitude range and frequency range of the digital signal that the target dual-mode intercom can meet when demodulating the digital signal, calibrate them as the standard amplitude range and standard frequency range, and in the target dual-mode intercom, respectively perform amplitude modulation and frequency modulation on the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal, so that the amplitude and frequency of the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal meet the standard amplitude range and standard frequency range respectively, and obtain the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

[0021] It should be noted that CNN is a neural network that is particularly suitable for processing image data. However, in speech recognition, CNN can also be used for feature extraction and classification by converting speech signals into image forms such as spectrograms. The convolution layer and pooling layer of CNN can automatically learn the local features and spatial relationships in the speech signal. Through the CNN neural network, the audio signal is segmented to obtain human voice and environmental sound. Because the audio that needs to be transmitted first is human voice, and the environmental sound needs to be transmitted. First, extract the features that can characterize the key information of the audio signal. The Mel spectrum analysis method can be used to extract features and convert the audio signal from the time domain to the frequency domain, so that it is easier to capture the difference in frequency between human voice and environmental sound. Build a CNN model suitable for audio segmentation tasks. The model may include multiple convolution layers, pooling layers, and fully connected layers. The convolution layer is used to extract local features in the audio signal, the pooling layer is used to reduce the dimension of the feature map and reduce the amount of calculation, and the fully connected layer is used to map the extracted features to the output category, that is, human voice and environmental sound, so according to the steps, the target denoised human voice audio signal and the target denoised environmental sound audio signal are obtained. After the audio is segmented, it is necessary to convert the audio into a digital signal, because the audio needs to be in the form of a digital signal to be transmitted. Adjusting the amplitude and frequency of the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal can realize the conversion of the audio signal into a digital signal, and obtain the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

[0022] Further, in a preferred embodiment of the present invention, the control target dual-mode intercom transmits the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and performs noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, specifically: Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, and calibrate the dual-mode intercom that receives the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal as the receiving dual-mode intercom; In the receiving dual-mode intercom, a target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal are received, and the received target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal are collectively referred to as received digital signals; In the receiving dual-mode intercom, adaptive differential pulse code modulation is performed on the received digital signal to decode the received digital signal into an audio signal to obtain a received audio signal, wherein the received audio signal includes a received target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal, which are calibrated as a received target noise reduction human voice audio signal and a received target noise reduction environmental sound audio signal; In the receiving dual-mode intercom, the received audio signal is converted into a spectrum to generate a received audio signal spectrum, a big data network is obtained, a standard frequency component range of the received audio signal is retrieved based on the big data network, and the received audio signal spectrum is analyzed. If there are frequency components that are not maintained within the standard frequency component range or there are clutters on the received audio signal spectrum, it is determined that the received audio signal has noise; If there is no frequency component that does not remain within the standard frequency component range on the received audio signal spectrum and no clutter exists, it is determined that there is no noise in the received audio signal; If there is no noise in the received audio signal, the receiving dual-mode intercom will give priority to outputting the receiving target noise reduction human voice audio signal, and determine whether the receiving target noise reduction ambient sound audio signal can be output in the receiving dual-mode intercom; If so, the received target noise reduction human voice audio signal and the received target noise reduction ambient sound audio signal are output simultaneously; if not, only the received target noise reduction human voice audio signal is output.

[0023] It should be noted that the receiving dual-mode intercom is determined, and the received digital signal is decoded and processed on the receiving dual-mode intercom, that is, adaptive differential pulse code modulation, so that the received digital signal is converted into a received audio signal to achieve the purpose of intercom audio transmission. During the transmission process, the audio may be affected by the environment or the equipment, which may cause the originally noise-free audio to continue to have noise. By performing frequency component analysis and clutter analysis on the signal spectrum of the received audio, it can be determined whether the received audio signal still has noise. If there is no noise, the noise-reduced human voice audio is output first, that is, the target noise-reduced human voice audio signal is received, because the intercom mainly transmits human voice audio. The environmental audio depends on the situation, and whether the environmental audio needs to be transmitted is manually controlled.

[0024] Figure 2 A flow chart of a method for tracing the noise source of a received audio signal and optimizing the communication quality is shown, comprising the following steps: S202: When there is noise in the received audio signal, it is necessary to trace the noise source of the received audio signal and optimize the communication quality in combination with the receiving dual-mode intercom and the surrounding environment; S204: performing communication frequency band optimization processing between the receiving dual-mode intercom and the target dual-mode intercom to optimize communication quality.

[0025] Furthermore, in a preferred embodiment of the present invention, when there is noise in the received audio signal, it is necessary to combine the receiving dual-mode intercom and the surrounding environment to trace the noise source of the received audio signal and optimize the communication quality, specifically: Obtain the communication range between the receiving dual-mode intercom and the target dual-mode intercom, and mark it as the target communication range; An environmental monitoring platform is determined in the big data network, and based on the environmental monitoring platform, environmental parameters in the target communication range are determined, calibrated as target communication range environmental parameters, and the real-time minimum communicable distance of the target noise-reduced human voice digital signal and the target noise-reduced environmental sound digital signal in the target communication range is measured, and the standard maximum communicable distance is determined at the same time; If the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal in the target communication range is greater than the standard maximum communicable distance, it is determined that the received audio signal has noise, which is an abnormality in the target communication range; Perform distance control on the target dual-mode intercom and the receiving dual-mode intercom, so that the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal is not greater than the standard maximum communicable distance, and continue to analyze the received audio signal spectrum; If there are still frequency components that are not maintained within the standard frequency component range on the received audio signal spectrum, or there are clutter, then the appropriate communication environment parameters of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal are retrieved in the big data network, and the scheme for adjusting the target communication range environment parameters so that the target communication range environment parameters are maintained equal to the appropriate communication environment parameters is retrieved and output; If there are still frequency components that are not maintained within the standard frequency component range or noise on the received audio signal spectrum, the communication frequency band optimization processing is performed between the receiving dual-mode walkie-talkie and the target dual-mode walkie-talkie to achieve communication quality optimization.

[0026] It should be noted that the presence of noise in the received audio signal indicates that the received audio signal is abnormal. The transmission distance of the audio and the environment during audio transmission will affect whether the audio generates noise. Therefore, it is necessary to analyze the communication distance during audio transmission to determine the relationship between the real-time minimum communication distance and the standard maximum communicable distance. If the real-time minimum communicable distance of the target noise-reduced human voice digital signal and the target noise-reduced ambient sound digital signal is greater than the standard maximum communicable distance, it is determined that the presence of noise in the received audio signal indicates that the target communication range is abnormal, and it is necessary to perform distance control of the intercom to shorten the communication distance. If there are still frequency components that are not maintained within the standard frequency component range after shortening the communication distance, or there are clutter, it is determined to be an environmental impact. At this time, it is necessary to directly retrieve the method output of the control 1 environmental parameters in the big data network.

[0027] Further, in a preferred embodiment of the present invention, the communication frequency band optimization processing is performed between the receiving dual-mode intercom and the target dual-mode intercom to achieve communication quality optimization, specifically: Determine the standard communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom, and check the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom to determine whether the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are greater than the standard communication frequency band; If not, the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are adjusted to make the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom equal to the standard communication frequency band; If the real-time communication frequency band of the receiving dual-mode intercom and the target dual-mode intercom is equal to the standard communication frequency band, but there are still frequency components that are not maintained within the standard frequency component range or noise in the received audio signal spectrum, the digital error correction technology is combined on the receiving dual-mode intercom to filter the noise of the received audio signal, and the signal of the received audio signal after the noise filtering is reconstructed, and finally the received audio signal after the signal reconstruction and noise filtering is output; The signal reconstruction is to adjust the frequency of the received audio signal after noise filtering so that the frequency of the received audio signal after noise filtering is equal to the frequency of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

[0028] It should be noted that when there are still frequency components that are not maintained within the standard frequency component range after the environmental parameters are adjusted, or there are clutter, it proves that the noise cannot be removed by adjusting the distance of the walkie-talkie and the environmental parameters. It is necessary to directly process the noise to improve the quality of the audio output. Walkie-talkies usually have multiple adjustable frequency bands or channels. Selecting a suitable frequency band is the first step to optimize the communication quality. Frequency bands that may interfere with other wireless communication devices or systems should be avoided. By reasonably allocating frequency bands and reducing frequency conflicts between different walkie-talkies, the reliability and stability of communication can be improved. Therefore, it is preferred to plan the communication frequency bands for the real-time communication frequency bands of the receiving dual-mode walkie-talkie and the target dual-mode walkie-talkie. If it still doesn't work, you can only use digital error correction technology, that is, directly output the audio signal with noise reduction, which means filtering the noise of the received audio signal, and reconstructing the received audio signal after the noise filtering, and finally outputting the received audio signal after signal reconstruction and noise filtering.

[0029] In addition, the communication quality intelligent tuning method applied to the dual-mode intercom also includes the following steps: The receiving dual-mode intercom and the target dual-mode intercom are collectively referred to as a dual-mode intercom system, and the communication time in the dual-mode intercom system is determined and calibrated as the target communication time; Introducing a dual-slot TDMA technology, based on which the target communication time is divided into time slots, so that the target communication time is divided into two time slots, which are marked as time slot 1 and time slot 2; Among them, different time slots can be used to send and receive audio signals; Control the operation of the dual-mode intercom system and maintain the synchronization of time slot 1 and time slot 2 during operation. If time slot 1 and time slot 2 are not synchronized, the dual-mode intercom system will pause when controlling the time slot with faster sending and receiving signals until time slot 1 and time slot 2 are not synchronized, then continue to control the operation of the two time slots.

[0030] It should be noted that dual-slot TDMA communication technology is a time division multiplexing technology that allows multiple users to communicate simultaneously in the same frequency band by allocating different time slots to different users or devices. In walkie-talkies, dual-slot TDMA technology can improve spectrum utilization, increase communication capacity, and reduce power consumption. In each time slot, the walkie-talkie can send or receive audio signals. Through reasonable time slot allocation, interference between different walkie-talkies can be avoided. In dual-slot TDMA communication, time slot synchronization is very important. If the time slots of the two walkie-talkies are not synchronized, communication failure or interference will occur. Therefore, an appropriate synchronization mechanism is needed to ensure the synchronization of time slots. Through the above method flow, dual-slot TDMA communication technology can be combined in two walkie-talkies to achieve noise control of transmitted audio. This can not only improve communication quality, but also reduce power consumption and cost, providing strong support for the widespread application of walkie-talkies.

[0031] like Figure 3 As shown, the second aspect of the present invention further provides a communication quality intelligent tuning system applied to a dual-mode walkie-talkie, the communication quality intelligent tuning system includes a memory 31 and a processor 32, the memory 31 stores a communication quality intelligent tuning method, and when the communication quality intelligent tuning method is executed by the processor 32, the following steps are implemented: Perform feature processing and feature analysis on the audio signal input into the dual-mode intercom to achieve noise reduction processing of the audio signal and obtain the target noise reduction audio signal; Performing audio feature analysis and audio segmentation on the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively; Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and perform noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal; When there is noise in the received audio signal, it is necessary to combine the receiving dual-mode walkie-talkie and the surrounding environment to trace the noise source and optimize the communication quality of the received audio signal.

[0032] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A communication quality intelligent tuning method applied to a dual-mode intercom, characterized in that: The following steps are involved: Perform feature processing and feature analysis on the audio signal input into the dual-mode intercom to achieve noise reduction processing of the audio signal and obtain the target noise reduction audio signal; Performing audio feature analysis and audio segmentation on the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively; Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and perform noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal; When there is noise in the received audio signal, it is necessary to combine the receiving dual-mode walkie-talkie and the surrounding environment to trace the noise source and optimize the communication quality of the received audio signal.

2. According to the method for intelligently optimizing the communication quality of a dual-mode intercom as described in claim 1, it is characterized in that: The feature processing and feature analysis of the audio signal input into the dual-mode intercom is performed to realize the noise reduction processing of the audio signal and obtain the target noise reduction audio signal, specifically: A dual-mode intercom that needs to be tuned for communication quality is obtained and calibrated as a target dual-mode intercom, wherein the target dual-mode intercom can collect audio and convert it into an audio signal, and perform encoding, decoding and feature analysis on the audio signal; Running the target dual-mode intercom to enable the target dual-mode intercom to collect audio signals in real time, and calibrating the audio signals as target audio signals, wherein the target audio signals include environmental audio signals and human voice audio signals; In the target dual-mode intercom, a target audio signal is subjected to DC component preprocessing and audio amplification to obtain a preprocessed target audio signal, the preprocessed target audio signal is subjected to Hanning window processing, and the preprocessed target audio signal is subjected to short-time Fourier transform after the Hanning window processing to generate a spectrum diagram of the preprocessed target audio signal, which is calibrated as a target spectrum diagram; In the target spectrum schematic diagram, quantile modeling analysis is performed on different frequency points on the target spectrum schematic diagram by using the quantile noise statistics method, and a quantile threshold is preset, and the frequency points whose quantiles are maintained within the quantile threshold are marked as noise signal frequency points, and the noise signal frequency points are marked and the target spectrum schematic diagram is updated to obtain an updated spectrum schematic diagram, wherein the updated spectrum schematic diagram records the spectrum characteristics of the noise signal; According to the spectral characteristics of the noise signal, the corresponding spectral flatness is calculated at the corresponding noise signal frequency point, wherein the spectral flatness is the difference between the noise spectral characteristics and the non-noise spectral characteristics at the noise signal frequency point, and based on the spectral flatness of different noise signal frequency points, the noise suppression coefficient of different noise signal frequency points is calculated by weighted average method; The noise suppression coefficients of different noise signal frequencies are imported into the updated spectrum diagram and spectrally multiplied with the corresponding noise signal frequencies to obtain the updated spectrum diagram after noise reduction. The updated spectrum diagram after noise reduction is subjected to an inverse short-time Fourier transform to obtain the target audio signal after noise reduction, which is calibrated as the target noise reduction audio signal.

3. According to the method for intelligently optimizing the communication quality of a dual-mode intercom as described in claim 1, it is characterized in that: The audio feature analysis and audio segmentation of the target noise reduction audio signal, and encoding the segmented audio signals into digital signals respectively, are specifically: Introducing a CNN neural network, wherein the CNN neural network includes a convolutional layer, a pooling layer and a fully connected layer; The sampling rate of the target noise reduction audio signal is adjusted to a sampling rate that the CNN neural network can perform sampling analysis on, thereby obtaining a sampled and adjusted noise reduction audio signal; Performing Mel spectrum conversion on the sampled and regulated noise-reduced audio signal to convert it into a sampled and regulated noise-reduced audio signal in the frequency domain state, importing the sampled and regulated noise-reduced audio signal in the frequency domain state into the convolution layer of the CNN neural network for convolution processing, and reducing the dimension of the sampled and regulated noise-reduced audio signal in the frequency domain state after the convolution processing through the pooling layer to realize local feature extraction of the sampled and regulated noise-reduced audio signal in the frequency domain state; Through the fully connected layer, the local features of the extracted sampled and regulated noise reduction audio signal are mapped to the human voice output category and the ambient sound output category to obtain the human voice audio signal and the ambient sound audio signal, which are calibrated as the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal; Determine the amplitude range and frequency range of the digital signal that the target dual-mode intercom can meet when demodulating the digital signal, calibrate them as the standard amplitude range and standard frequency range, and in the target dual-mode intercom, respectively perform amplitude modulation and frequency modulation on the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal, so that the amplitude and frequency of the target noise reduction human voice audio signal and the target noise reduction ambient sound audio signal meet the standard amplitude range and standard frequency range respectively, and obtain the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

4. According to the method for intelligently optimizing the communication quality of a dual-mode intercom as described in claim 1, it is characterized in that: The control target dual-mode intercom transmits the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal to other dual-mode intercoms for decoding, and performs noise analysis on the decoded target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, specifically: Control the target dual-mode intercom to transmit the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal, and calibrate the dual-mode intercom that receives the target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal as the receiving dual-mode intercom; In the receiving dual-mode intercom, a target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal are received, and the received target noise reduction human voice digital signal and the target noise reduction environmental sound digital signal are collectively referred to as received digital signals; In the receiving dual-mode intercom, adaptive differential pulse code modulation is performed on the received digital signal to decode the received digital signal into an audio signal to obtain a received audio signal, wherein the received audio signal includes a received target noise reduction human voice digital signal and a target noise reduction environmental sound digital signal, which are calibrated as a received target noise reduction human voice audio signal and a received target noise reduction environmental sound audio signal; In the receiving dual-mode intercom, the received audio signal is converted into a spectrum to generate a received audio signal spectrum, a big data network is obtained, a standard frequency component range of the received audio signal is retrieved based on the big data network, and the received audio signal spectrum is analyzed. If there are frequency components that are not maintained within the standard frequency component range or there are clutters on the received audio signal spectrum, it is determined that the received audio signal has noise; If there is no frequency component that does not remain within the standard frequency component range on the received audio signal spectrum and no clutter exists, it is determined that there is no noise in the received audio signal; If there is no noise in the received audio signal, the receiving dual-mode intercom will give priority to outputting the receiving target noise reduction human voice audio signal, and determine whether the receiving target noise reduction ambient sound audio signal can be output in the receiving dual-mode intercom; If so, the received target noise reduction human voice audio signal and the received target noise reduction ambient sound audio signal are output simultaneously; if not, only the received target noise reduction human voice audio signal is output.

5. According to the method for intelligently optimizing the communication quality of a dual-mode intercom as described in claim 1, it is characterized in that: When the received audio signal has noise, it is necessary to combine the receiving dual-mode intercom and the surrounding environment to trace the noise source and optimize the communication quality of the received audio signal, specifically: Obtain the communication range between the receiving dual-mode intercom and the target dual-mode intercom, and mark it as the target communication range; An environmental monitoring platform is determined in the big data network, and based on the environmental monitoring platform, environmental parameters in the target communication range are determined, calibrated as target communication range environmental parameters, and the real-time minimum communicable distance of the target noise-reduced human voice digital signal and the target noise-reduced environmental sound digital signal in the target communication range is measured, and the standard maximum communicable distance is determined at the same time; If the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal in the target communication range is greater than the standard maximum communicable distance, it is determined that the received audio signal has noise, which is an abnormality in the target communication range; Perform distance control on the target dual-mode intercom and the receiving dual-mode intercom, so that the real-time minimum communicable distance of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal is not greater than the standard maximum communicable distance, and continue to analyze the received audio signal spectrum; If there are still frequency components that are not maintained within the standard frequency component range on the received audio signal spectrum, or there are clutter, then the appropriate communication environment parameters of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal are retrieved in the big data network, and the scheme for adjusting the target communication range environment parameters so that the target communication range environment parameters are maintained equal to the appropriate communication environment parameters is retrieved and output; If there are still frequency components that are not maintained within the standard frequency component range or noise on the received audio signal spectrum, the communication frequency band optimization processing is performed between the receiving dual-mode walkie-talkie and the target dual-mode walkie-talkie to achieve communication quality optimization.

6. According to the method for intelligently optimizing the communication quality of a dual-mode intercom as described in claim 5, it is characterized in that: The communication frequency band optimization process is performed between the receiving dual-mode intercom and the target dual-mode intercom to achieve communication quality optimization, specifically: Determine the standard communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom, and check the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom to determine whether the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are greater than the standard communication frequency band; If not, the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom are adjusted to make the real-time communication frequency bands of the receiving dual-mode intercom and the target dual-mode intercom equal to the standard communication frequency band; If the real-time communication frequency band of the receiving dual-mode intercom and the target dual-mode intercom is equal to the standard communication frequency band, but there are still frequency components that are not maintained within the standard frequency component range or noise in the received audio signal spectrum, the digital error correction technology is combined on the receiving dual-mode intercom to filter the noise of the received audio signal, and the signal of the received audio signal after the noise filtering is reconstructed, and finally the received audio signal after the signal reconstruction and noise filtering is output; The signal reconstruction is to adjust the frequency of the received audio signal after noise filtering so that the frequency of the received audio signal after noise filtering is equal to the frequency of the target noise reduction human voice digital signal and the target noise reduction ambient sound digital signal.

7. A communication quality intelligent tuning system applied to dual-mode intercoms, characterized in that: The communication quality intelligent tuning system includes a memory and a processor, wherein a communication quality intelligent tuning method program is stored in the memory, and when the communication quality intelligent tuning method program is executed by the processor, the communication quality intelligent tuning method steps as described in any one of claims 1-6 are implemented.

Images