A method, system, device, and medium for rapid interpretation of flight parameter audio data.

Abstract

The application provides a kind of fast interpretation method, system, equipment and medium of flight parameter audio data, method includes: to original flight parameter audio data decryption obtains the single-channel audio data packet of multiple channels;Using 711 encoding and decoding algorithm single-channel audio data packet decoding, using high-pass filter algorithm is carried out high noise signal data separation to the data after decoding, obtains voice data set and abnormal audio data set;Using the trained voice recognition model is carried out semantic recognition to voice data set and obtains playable format audio file;Using the aircraft abnormal sound model after training is carried out once interpretation to the abnormal audio data set;Using flight parameter audio semantic interpretation library, the playable format audio file is carried out secondary interpretation;Fusion twice interpretation results, output whether there is fault, fault type and fault occurrence time.The application can be highly automated, improve aircraft fault analysis efficiency, can improve ground maintenance efficiency.

Description

Technical Field

[0001] This invention belongs to the field of avionics and relates to audio data judgment technology, specifically to a method, system, device and medium for rapid interpretation of flight parameter audio data. Background Technology

[0002] Currently, the traditional method for interpreting flight parameter audio data is to convert the data format using audio playback software and then play it back, followed by manual interpretation of the played-back aircraft audio. This method suffers from long interpretation times and low efficiency in extracting useful information.

[0003] Furthermore, with the rapid development of information technology, while continuously advancing the application and analysis of flight data, attention is also being paid to the interpretation and analysis of audio data. Because flight parameter audio data has certain industry-specific characteristics, and rapid interpretation of flight parameter voice data is a direction with industry demand, research is needed on technologies that enable flight parameter voice data interpretation. Summary of the Invention

[0004] To address the technical problems of long interpretation time and low efficiency in extracting useful information when manually listening to and interpreting aircraft audio data, this invention provides a rapid interpretation method for flight parameter audio data. The rapid interpretation method includes the following steps:

[0005] S1. Decrypt the raw flight parameter audio data read into memory to obtain single-channel audio data packets for each channel;

[0006] S2. The 711 encoding and decoding algorithm is used to decode all the single-channel audio data packets respectively. The high-pass filtering algorithm is used to separate the high noise signal data of the decoded data to obtain the voice dataset and the abnormal audio dataset. The abnormal audio dataset includes the high noise signal data of each channel.

[0007] S3. Use the trained speech recognition model to perform semantic recognition on the speech dataset, and store the speech information and time information of each channel as a playable audio file.

[0008] S4. Using the trained aircraft abnormal sound model, perform a judgment on the abnormal audio dataset and output the judgment result.

[0009] S5. Using the flight parameter audio semantic interpretation library, perform secondary interpretation on the playable audio file to obtain the secondary interpretation result;

[0010] S6. Combine the first reading result and the second reading result to obtain the reading result, which includes whether there is a fault, the type of fault, and the time of fault occurrence.

[0011] Further, in step S1 above, the decryption of the original flight parameter audio data read into memory to obtain a single-channel audio data packet for each channel includes:

[0012] S11. For the raw flight parameter audio data collected by the recorder, identify the raw audio data packet of each channel according to the file extension and history file, and read the raw audio data packet of each channel into the memory;

[0013] S12. Using the protocol corresponding to the audio format of each channel, decrypt each of the original audio data packets in the memory to obtain a single-channel audio data packet corresponding to each channel.

[0014] Furthermore, in step S12 above, the protocol is the 65A protocol, and the single-channel audio data packet is a single-channel DAN format audio data packet.

[0015] Furthermore, in step S2 above, before using the high-pass filtering algorithm to separate high-noise signal data from the decoded data, the following steps are also included:

[0016] A low-pass filtering algorithm is used to filter low-noise signals for each decoded data.

[0017] Furthermore, in step S3 above, the playable audio file format includes .DAT format audio files or .WAV format audio files.

[0018] In an improved embodiment of the fast interpretation method, the method further includes constructing a trained aircraft anomaly sound model, specifically comprising the following steps:

[0019] The historical abnormal audio dataset is divided into a training set and a test set. The training set is transformed and analyzed using a wavelet transform algorithm. Sound features are extracted by calculating the Mel frequency cepstral coefficients to obtain MFCC sound feature data.

[0020] An initial model of abnormal aircraft sounds was constructed by iteratively training the MFCC sound feature data using an SVM classifier.

[0021] The initial model of abnormal aircraft sounds is optimized using the test set to obtain the trained model of abnormal aircraft sounds.

[0022] In another improved embodiment of the rapid interpretation method, a flight parameter audio semantic interpretation library is also included. The specific method is as follows: common aircraft faults are classified, semantic criteria are added based on expert experience, and a flight parameter audio semantic interpretation library is constructed.

[0023] This invention also provides a rapid interpretation system for flight parameter audio data, including a reading and parsing module, a decoding module, a filtering module, a semantic recognition and conversion module, a first interpretation module, a second interpretation module, and a fusion module.

[0024] The read / decrypt module is used to decrypt the original flight parameter audio data read into memory to obtain single-channel audio data packets for each channel.

[0025] The decoding module is used to decode all the single-channel audio data packets using the 711 encoding / decoding algorithm;

[0026] The filtering module is used to separate high-noise signal data from the decoded data using a high-pass filtering algorithm to obtain a voice dataset and an abnormal audio dataset. The abnormal audio dataset includes high-noise signal data for each of the channels.

[0027] The semantic recognition and conversion module is used to perform semantic recognition on the speech dataset using a trained speech recognition model, and to store the speech information and time information of each channel as a playable audio file.

[0028] The first judgment module is used to perform a judgment on the abnormal audio dataset using the trained aircraft abnormal sound model and output a judgment result.

[0029] The second interpretation module is used to perform secondary interpretation on the playable audio file using the flight parameter audio semantic interpretation library to obtain the secondary interpretation result;

[0030] The fusion module is used to fuse the first judgment result and the second judgment result to obtain a judgment result, which includes whether there is a fault, the fault type, and the fault occurrence time.

[0031] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned method for quickly interpreting any flight parameter audio data, thereby solving the technical problems of long interpretation time and low efficiency of useful information extraction in the prior art, which relies on manually listening to the playback of aircraft voice for interpretation.

[0032] This invention also provides a computer-readable storage medium storing a computer program that executes the above-described method for rapidly interpreting any of the flight parameter audio data. This solves the technical problems of long interpretation time and low efficiency in extracting useful information in the prior art, which relies on manually listening to the played-back aircraft voice for interpretation.

[0033] Compared with existing technologies, the beneficial effects achieved by at least one of the above-mentioned technical solutions adopted in the embodiments of this specification include at least the following: The rapid interpretation method for flight parameter audio data of the present invention separates the voice dataset and the abnormal audio dataset from the original flight parameter audio data, performs semantic recognition on the abnormal audio dataset using a trained aircraft abnormal sound model, performs interpretation on the voice dataset using a flight parameter audio semantic interpretation library, and outputs the final interpretation result after fusing the two interpretation results. The above-mentioned method of the present invention can shorten the ground maintenance support time for aircraft, open up a new interpretation approach for flight parameter audio, effectively improve the efficiency of aircraft fault analysis, can be applied to the general platform of aircraft support equipment, and is not limited by aircraft model, thus possessing universality. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of a method for rapidly interpreting flight parameter audio data disclosed in an embodiment of the present invention;

[0036] Figure 2 This is a flowchart illustrating the rapid interpretation of flight parameter audio data as disclosed in an embodiment of the present invention.

[0037] Figure 3 This is an architecture diagram of the rapid interpretation system for flight parameter audio data disclosed in an embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram of a computer device disclosed in an embodiment of the present invention;

[0039] Among them, 301 is the reading and parsing module; 302 is the decoding module; 303 is the filtering module; 304 is the semantic recognition and conversion module; 305 is the first judgment module; 306 is the second judgment module; 307 is the fusion module; 401 is the memory; and 402 is the processor. Detailed Implementation

[0040] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0041] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features of the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] This invention discloses a method for rapid interpretation of flight parameter audio data. It uses flight parameter audio data recorded by an aircraft recorder as the object, analyzes and denoises the aircraft voice data, extracts the voices of multiple pilots from the voice data, identifies and obtains the voice information contained in the voice data, the time of occurrence of the voices, and extracts abnormal sound information from the voice data. Through audio data fusion, it automatically outputs interpretation information. The following is a detailed description of the above-mentioned rapid interpretation method of this invention:

[0043] See Figure 1 and Figure 2 As shown, the method for rapid interpretation of flight parameter audio data includes the following steps:

[0044] S1. Decrypt the raw flight parameter audio data read into memory to obtain single-channel audio data packets for each channel;

[0045] S2. The 711 encoding and decoding algorithm is used to decode all the single-channel audio data packets respectively. The high-pass filtering algorithm is used to separate the high noise signal data of the decoded data to obtain the voice dataset and the abnormal audio dataset. The abnormal audio dataset includes the high noise signal data of each channel.

[0046] S3. Use the trained speech recognition model to perform semantic recognition on the speech dataset, and store the speech information and time information of each channel as a playable audio file.

[0047] S4. Using the trained aircraft abnormal sound model, perform a judgment on the abnormal audio dataset and output the judgment result.

[0048] S5. Using the flight parameter audio semantic interpretation library, perform secondary interpretation on the playable audio file to obtain the secondary interpretation result;

[0049] S6. Combine the first reading result and the second reading result to obtain the reading result, which includes whether there is a fault, the type of fault, and the time of fault occurrence.

[0050] Further, in step S1 above, the decryption of the original flight parameter audio data read into memory to obtain a single-channel audio data packet for each channel includes:

[0051] S11. For the raw flight parameter audio data collected by the recorder, identify the raw audio data packet of each channel according to the file extension and history file, and read the raw audio data packet of each channel into the memory;

[0052] S12. Using the protocol corresponding to the audio format of each channel, decrypt each of the original audio data packets in the memory to obtain a single-channel audio data packet corresponding to each channel.

[0053] Furthermore, in step S12 above, the protocol is the 65A protocol, and the single-channel audio data packet is a single-channel DAN format audio data packet.

[0054] Furthermore, in step S2 above, before using the high-pass filtering algorithm to separate high-noise signal data from the decoded data, see [link to previous section]. Figure 2 As shown, it also includes:

[0055] A low-pass filtering algorithm is used to filter low-noise signals in each decoded data. By filtering out low-noise signals in the audio, the accuracy of subsequent interpretation results can be improved.

[0056] Furthermore, in step S3 above, the playable audio file format includes .DAT format audio files or .WAV format audio files.

[0057] In an improved embodiment of the fast interpretation method, the method further includes constructing a trained aircraft anomaly sound model, specifically comprising the following steps:

[0058] The historical abnormal audio dataset is divided into a training set and a test set. The training set is transformed and analyzed using a wavelet transform algorithm. Sound features are extracted by calculating the Mel frequency cepstral coefficients to obtain MFCC sound feature data.

[0059] An initial model of abnormal aircraft sounds was constructed by iteratively training the MFCC sound feature data using an SVM classifier.

[0060] The initial model of abnormal aircraft sounds is optimized using the test set to obtain the trained model of abnormal aircraft sounds.

[0061] In another improved embodiment of the rapid interpretation method, a flight parameter audio semantic interpretation library is also included. The specific method is as follows: common aircraft faults are classified, semantic criteria are added based on expert experience, and a flight parameter audio semantic interpretation library is constructed.

[0062] Based on the same inventive concept, this invention also provides a rapid interpretation system for flight parameter audio data, as described in the following embodiments. Since the principle of the rapid interpretation system in solving the problem is similar to that of the rapid interpretation method, the implementation of the rapid interpretation system can refer to the implementation of the aforementioned rapid interpretation method, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0063] Figure 3 This is a structural block diagram of a rapid interpretation system for flight parameter audio data according to an embodiment of the present invention, such as... Figure 3 As shown, the fast interpretation system for flight parameter audio data includes a reading and parsing module 301, a decoding module 302, a filtering module 303, a semantic recognition and conversion module 304, a first interpretation module 305, a second interpretation module 306, and a fusion module 307. The structure is described below.

[0064] The read / decryption module 301 is used to decrypt the original flight parameter audio data read into the memory to obtain a single-channel audio data packet for each channel.

[0065] The decoding module 302 is used to decode all the single-channel audio data packets using the 711 encoding and decoding algorithm;

[0066] The filtering module 303 is used to separate high-noise signal data from the decoded data using a high-pass filtering algorithm to obtain a voice dataset and an abnormal audio dataset. The abnormal audio dataset includes high-noise signal data for each of the channels.

[0067] The semantic recognition and conversion module 304 is used to perform semantic recognition on the speech dataset using a trained speech recognition model, and to store the speech information and time information of each channel as a playable audio file.

[0068] The first judgment module 305 is used to use the trained aircraft abnormal sound model to perform a judgment on the abnormal audio dataset and output a judgment result.

[0069] The second interpretation module 306 is used to perform secondary interpretation on the playable audio file using the flight parameter audio semantic interpretation library to obtain the secondary interpretation result;

[0070] The fusion module 307 is used to fuse the first judgment result and the second judgment result to obtain a judgment result, which includes whether there is a fault, the fault type, and the fault occurrence time.

[0071] The embodiments of this invention achieve the following technical effects: The rapid interpretation method for flight parameter audio data of this invention separates the voice dataset and the abnormal audio dataset from the original flight parameter audio data. It then uses a trained aircraft abnormal sound model to interpret the abnormal audio dataset, and uses a flight parameter audio semantic interpretation library to perform semantic recognition on the voice dataset. Finally, it fuses the two interpretation results to output the final interpretation result. This method can shorten the ground maintenance and support time for aircraft, opens up a new interpretation approach for flight parameter audio, effectively improves the efficiency of aircraft fault analysis, and can be applied to a general platform for aircraft support equipment without being limited by aircraft model, thus possessing versatility.

[0072] In this embodiment, a computer device is provided, such as... Figure 4 As shown, it includes a memory 401, a processor 402, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned method for quickly interpreting any flight parameter audio data.

[0073] Specifically, the computer device can be a computer terminal, a server, or a similar computing device.

[0074] In this embodiment, a computer-readable storage medium is provided, which stores a computer program that executes the fast interpretation method for any of the flight parameter audio data described above.

[0075] Specifically, computer-readable storage media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. As defined herein, computer-readable storage media do not include transient computer-readable media, such as modulated data signals and carrier waves.

[0076] Obviously, those skilled in the art should understand that the modules or steps of the above-described embodiments of the present invention can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the embodiments of the present invention are not limited to any particular hardware and software combination.

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations can be made to the embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for fast interpretation of flight parameter audio data, characterized in that, The method comprises the following steps: decrypting the original flight parameter audio data read into the memory to obtain single-channel audio data packets of each channel; decoding all the single-channel audio data packets by using a 711 coding and decoding algorithm, separating high-noise signal data from the decoded data by using a high-pass filtering algorithm to obtain a voice data set and an abnormal audio data set, the abnormal audio data set comprising high-noise signal data of each channel; performing semantic recognition on the voice data set by using a trained voice recognition model, and storing the voice information of each channel and its time information in a playable format audio file; performing a first interpretation on the abnormal audio data set by using a trained aircraft abnormal sound model, and outputting a first interpretation result; performing a second interpretation on the playable format audio file by using a flight parameter audio semantic interpretation library to obtain a second interpretation result; fusing the first interpretation result and the second interpretation result to obtain an interpretation result, the interpretation result comprising whether there is a fault, a fault type and a fault occurrence time.

2. The method of rapid interpretation of flight parameter audio data of claim 1, wherein, The step of decrypting the original flight parameter audio data read into the memory to obtain single-channel audio data packets of each channel comprises the following steps: identifying the original audio data packets of each channel according to the file suffix name and the history file of the original flight parameter audio data collected by the recorder, and reading the original audio data packets of each channel into the memory; decrypting each original audio data packet in the memory by using a protocol corresponding to the audio format of each channel to obtain single-channel audio data packets corresponding to each channel.

3. The method of claim 2, wherein, The protocol is a 65A protocol, and the single-channel audio data packets are single-channel DAN format audio data packets.

4. The method of claim 1, wherein, Before separating high-noise signal data from the decoded data by using a high-pass filtering algorithm, the method further comprises the following step: filtering low-noise signals from each decoded data by using a low-pass filtering algorithm.

5. The method of claim 1, wherein, The playable format audio file comprises a.DAT format audio file or a.WAV format audio file.

6. The method of rapid interpretation of flight parameter audio data of claim 1, wherein, The method further comprises the following steps: dividing a historical abnormal audio data set into a training set and a test set, transforming and analyzing the training set by using a wavelet transform algorithm, extracting sound features by calculating mel frequency cepstral coefficients to obtain MFCC sound feature data; iteratively training the MFCC sound feature data by using an SVM classifier to construct an initial aircraft abnormal sound model; optimizing the initial aircraft abnormal sound model by using the test set to obtain the trained aircraft abnormal sound model.

7. The method of rapid interpretation of flight parameter audio data of claim 1, wherein, The method further comprises the following steps: classifying common aircraft faults, adding semantic criteria according to expert experience, and constructing a flight parameter audio semantic interpretation library.

8. A system for rapid interpretation of acoustic data from a flight, characterized by The method comprises the following steps: a reading and analyzing module for decrypting original flight parameter audio data read into the memory to obtain single-channel audio data packets of each channel; a decoding module for decoding all the single-channel audio data packets by using a 711 coding and decoding algorithm; a filtering module, configured to separate high-noise signal data from the decoded data by using a high-pass filtering algorithm, to obtain a voice data set and an abnormal audio data set, the abnormal audio data set including high-noise signal data of each channel; a semantic recognition and conversion module, configured to recognize semantics of the voice data set by using a trained voice recognition model, and store voice information and time information of each channel obtained by recognition in a playable format audio file; a first interpretation module, configured to interpret the abnormal audio data set by using a trained aircraft abnormal sound model, and output a first interpretation result; a second interpretation module, configured to interpret the playable format audio file by using a flight parameter audio semantic interpretation library, and obtain a second interpretation result; a fusion module, configured to fuse the first interpretation result and the second interpretation result, and obtain an interpretation result, the interpretation result including whether there is a fault, a fault type, and a fault occurrence time.

9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method for rapid interpretation of flight parameter audio data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program for implementing the method for rapid interpretation of flight parameter audio data according to any one of claims 1 to 7.

Images