Audio signal quality assessment method and device

By calculating the evaluation parameters and number of lags of audio packet data, the audio signal quality evaluation without reference is achieved, which solves the problem of inaccurate and real-time audio signal quality evaluation in the prior art, and improves audio signal quality monitoring in audio and video calls and evaluate efficiency.

CN118658497BActive Publication Date: 2024-11-15BEIJING SMARTER EYE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411131803.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-15
Estimated Expiration
2044-08-19

AI Technical Summary

Technical Problem

The prior art cannot achieve referenceless evaluation of audio signal quality, resulting in inaccurate and real-time evaluation, and the problem of sound lag and sound quality changes cannot be quickly positioned and solved in audio and video call scenarios.

Method used

By obtaining audio packet data, calculate the evaluation parameters of the audio packet data, such as packet loss rate, jitter value, packet loss compensation value, quick broadcast data and stretch data, determine the number of audio stuttering based on the preset conditions, and determine the number of stuttering based on the number of stuttering times, and determine the number of stuttering times based on the number of stuttering times. The mapping relationship with the evaluation score generates evaluation scores to achieve reference-free audio signal quality evaluation.

Benefits of technology

It improves the convenience, accuracy and real-time performance of audio signal quality evaluation, can timely locate problems during audio and video calls, save system repair time, and improve the effect and stability of the overall audio and video system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118658497B_ABST
    Figure CN118658497B_ABST
Patent Text Reader

Abstract

The present invention provides an audio signal quality assessment method and device, the method comprising: obtaining audio packet data of a target audio to be assessed, calculating assessment parameters of the audio packet data; determining that audio freeze occurs at the current moment in the target audio when the assessment parameters meet preset assessment conditions; calculating the number of audio freezes within the assessment duration; and generating an assessment score for the target audio based on a mapping relationship between the number of freezes and the assessment score that is preset. After receiving the audio packet information, audio packet loss calculation, audio jitter value calculation, PLC packet loss compensation calculation, audio quick broadcast information calculation, and audio stretching information calculation are performed, and then the final result is obtained based on the calculation results and preset strategies. Thereby, a reference-free assessment of the audio signal quality is achieved, and the convenience, accuracy, and real-time performance of the audio signal quality assessment are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of signal quality assessment, and in particular to an audio signal quality assessment method and device. Background Art

[0002] At present, in audio and video calls or audio and video conference application scenarios, due to the diversity of audio and video acquisition devices (including Windows, Android, iPhone, Mac, etc. and various external devices), the diversity of network environments, audio and video encoding algorithms, audio and video processing algorithms, and the transmission process of audio and video signals, it may cause sound freezes and changes in sound quality, causing discomfort to people's hearing. So how to measure and evaluate the quality of audio signals after loss has become an important research direction. Accurate, real-time, and convenient audio-based transmission signal quality assessment can not only enable professionals to quickly locate problems, but also provide references for providing better audio signal quality and provide data support for improving signal transmission quality.

[0003] In the existing technology, the audio signal evaluation method is mainly PESQ (Perceptual evaluation of speech quality). PESQ is an objective, full-reference audio signal quality evaluation method. The PESQ algorithm requires a noisy attenuated signal and an original reference signal. It can provide a subjective MOS prediction value for the objective audio signal quality evaluation and can be mapped to the MOS scale range. The PESQ score range is between 0.5 and 4.5. The higher the score, the better the speech quality. However, this method is a full-reference method. It relies on the comparison between the original signal and the processed signal and cannot achieve no-reference evaluation.

[0004] In view of this, providing an audio signal quality assessment method and device to achieve reference-free assessment of audio signal quality has become an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] The present invention provides an audio signal quality assessment method and device to achieve reference-free assessment of audio signal quality, thereby improving the convenience, accuracy and real-time performance of audio signal quality assessment.

[0006] The present invention provides an audio signal quality assessment method, the method comprising:

[0007] Obtain audio packet data of the target audio to be evaluated;

[0008] Calculating evaluation parameters of the audio packet data;

[0009] When the evaluation parameter satisfies a preset evaluation condition, determining that audio freeze occurs in the target audio at the current moment;

[0010] Calculate the number of audio freezes that occur during the evaluation period;

[0011] The evaluation score of the target audio is generated according to a mapping relationship between the preset number of freezes and the evaluation score.

[0012] In some embodiments, the evaluation parameters of the audio packet data include an audio packet loss rate, an audio jitter value, a packet loss compensation value, audio quick play data, and audio stretching data.

[0013] In some embodiments, the audio jitter value is calculated using a first expression , the first expression is:

[0014]

[0015] in, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet, Representative The sending timestamp of each data packet, Representative The receiving timestamp of the data packet.

[0016] In some embodiments, calculating the audio quick play data and the audio stretching data specifically includes:

[0017] Decode the audio data packet to obtain an audio PCM signal;

[0018] Combined with the voice control speed change factor, determine whether the input audio PCM signal needs speed change processing;

[0019] When it is determined that the audio PCM signal needs speed change processing, short-time energy estimation and similarity test are performed on the audio frames of the audio PCM signal, and waveform superposition is performed through the quality control factor to obtain an output voice signal;

[0020] When it is determined that the audio PCM signal does not need speed change processing, the input audio PCM signal is used as the output voice signal.

[0021] In some embodiments, the audio freeze includes major freeze and minor freeze; when the evaluation parameter satisfies a preset evaluation condition, determining that the target audio has audio freeze at the current moment specifically includes:

[0022] When the evaluation parameter satisfies a preset first evaluation condition, determining that a major freeze occurs in the target audio at the current moment;

[0023] When the evaluation parameter satisfies a preset second evaluation condition, it is determined that a slight freeze occurs in the target audio at the current moment.

[0024] In some embodiments, the first evaluation condition includes at least one of the following:

[0025] Within the preset time period, the total number of active packet losses and passive packet losses exceeds the first preset time period;

[0026] In two consecutive preset time periods, the number of active packet losses and passive packet losses in each preset time period exceeds the second preset time period;

[0027] Within the preset time, the number of consecutive packet loss compensations exceeds the preset compensation times.

[0028] Within the preset time, the data lost by fast playback exceeds the first preset time;

[0029] Within the preset time period, the data expanded by stretching exceeds a third preset time period;

[0030] Within the preset time, the packet loss compensation process exceeds the preset compensation process times;

[0031] Within the preset time length, the average audio jitter value is greater than a fourth preset time length.

[0032] The present invention also provides an audio signal quality assessment device, the device comprising:

[0033] A data acquisition unit, used to acquire audio packet data of the target audio to be evaluated;

[0034] a parameter calculation unit, configured to calculate evaluation parameters of the audio packet data;

[0035] A jam calculation unit, configured to determine that audio jam occurs in the target audio at a current moment when the evaluation parameter satisfies a preset evaluation condition, and calculate the number of times the audio jam occurs within an evaluation time period;

[0036] A result generating unit is used to generate an evaluation score of the target audio according to a preset mapping relationship between the number of freezes and the evaluation score.

[0037] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described above when executing the program.

[0038] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method described above when executed by a processor.

[0039] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method described above is implemented.

[0040] The audio signal quality assessment method and device provided by the present invention obtain the audio packet data of the target audio to be assessed, calculate the assessment parameters of the audio packet data; when the assessment parameters meet the preset assessment conditions, determine whether the target audio has audio freeze at the current moment; calculate the number of audio freezes within the assessment time; and generate the assessment score of the target audio according to the mapping relationship between the preset number of freezes and the assessment score. After receiving the audio packet information, audio packet loss calculation, audio jitter value calculation, PLC packet loss compensation calculation, audio fast broadcast information calculation and audio stretching information calculation are performed, and then the final result is obtained according to the calculation results and the preset strategy. Thereby, the reference-free assessment of the audio signal quality is realized, and the convenience, accuracy and real-time performance of the audio signal quality assessment are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 is one of the flow charts of the audio signal quality assessment method provided by the present invention;

[0043] Figure 2 This is the second flowchart of the audio signal quality assessment method provided by the present invention;

[0044] Figure 3 It is a schematic diagram of the principle of audio jitter provided by the present invention;

[0045] Figure 4 This is a diagram showing the effect of the voice signal after packet loss compensation provided by the present invention;

[0046] Figure 5 It is a flow chart of the audio quick play algorithm and the stretching algorithm provided by the present invention;

[0047] Figure 6 is a structural block diagram of an audio signal quality assessment device provided by the present invention;

[0048] Figure 7It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Among the existing audio signal quality evaluation methods, the PESQ-based evaluation method is a full-reference method and cannot achieve non-reference evaluation; the evaluation method based on delay, packet loss rate, and network jitter value is relatively cumbersome, not intuitive, and not accurate enough; the method based on subjective evaluation requires high manpower and time costs and cannot be evaluated in real time.

[0051] In view of the shortcomings and deficiencies in the existing technology, combined with active packet loss, passive packet loss, PLC (Packets Loss Cancellation) packet loss compensation algorithm, audio data fast broadcast algorithm, audio data stretching algorithm, audio jitter value, etc., according to certain strategies and algorithms, an audio signal quality evaluation method is provided to monitor and evaluate the audio signal quality in real time in the background, so that the audio problem at a certain moment during the call can be located in time and quickly, or the audio signal quality of the entire call process can be evaluated, which greatly saves the time to repair system problems and greatly improves the effect and stability of the overall audio and video system.

[0052] In a specific embodiment, Figure 1 As shown, the audio signal quality assessment method provided by the present invention comprises the following steps:

[0053] S110: Acquire audio packet data of target audio to be evaluated; the target audio is call audio, audio and video call, or audio and video conference to be evaluated.

[0054] S120: Calculate evaluation parameters of the audio packet data; wherein the evaluation parameters of the audio packet data include audio packet loss rate, audio jitter value, packet loss compensation value, audio quick play data and audio stretching data.

[0055] S130: When the evaluation parameter satisfies a preset evaluation condition, determine whether audio freeze occurs in the target audio at the current moment; for example, whether audio freeze occurs within 200 ms or within 100 ms.

[0056] S140: Calculate the number of times audio freezes occur within the evaluation time; for example, the number of times audio freezes occur within 200ms is 2 times, etc.

[0057] S150: Generate an evaluation score for the target audio according to a pre-set mapping relationship between the number of freezes and the evaluation score. For example, taking a 5-point system as an example, if the number of freezes is counted to be 4 within 200ms, the corresponding evaluation score is 3 points, if the number of freezes is counted to be 2 within 200ms, the corresponding evaluation score is 4 points, and so on.

[0058] It is understandable that the present invention provides a reference-free audio signal quality assessment method that does not require other additional reference audio files or technical parameters, etc. Compared with the existing audio signal assessment scheme, it is more concise, convenient, more accurate, and saves space and manpower. Furthermore, the quality assessment results can be displayed in real time through the background in the form of assessment scores, which is more real-time and more intuitive.

[0059] In the overall process, Figure 2 As shown, the audio signal quality assessment method is arranged at the receiving end in the form of an audio signal quality assessment module. After receiving the audio packet information, audio packet loss calculation, audio jitter value calculation, PLC packet loss compensation calculation, audio quick broadcast information calculation and audio stretching information calculation are performed, and then these information are sent to the audio signal quality assessment module to obtain the final result according to certain strategies and algorithms. In this way, through the reference-free audio signal quality assessment method proposed in the present invention, the audio signal quality during audio and video calls can be evaluated in real time, accurately and conveniently. Compared with the existing audio signal quality assessment method, it not only improves the real-time performance and can monitor the audio signal quality in real time, but also improves the accuracy and intuitiveness, and can intuitively draw conclusions through background data, while saving a lot of manpower and time costs.

[0060] like Figure 3 As shown, Sender is the sender, Receiver is the receiver, t represents the time axis, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet is defined as , then use the first expression to calculate the audio jitter value , the first expression is:

[0061] .

[0062] When calculating audio packet loss, it includes active packet loss and passive packet loss. Passive packet loss means that when audio data passes through the network from the sender A to the receiver B, the data packet may be lost in the network due to link fluctuations, router cache queue backlog, or poor Wi-Fi signal. These packet losses are all passive packet losses. In a set period, the ratio of the total number of packets lost to the theoretical number of packets to be received is the packet loss rate. Active packet loss refers to the active abandonment of a portion of audio data packets. In the audio network jitter processing module, since the audio and video call system is a real-time system, the cache cannot be set too large. When the network environment is poor and the cached audio data packets are greater than a certain threshold, a portion of the audio data packets must be actively abandoned. This portion of audio data packets is called active packet loss.

[0063] The audio packet loss compensation algorithm PLC, the full name is Packets Loss Cancellation, which is a packet loss concealment technology. During audio data transmission, when audio packet loss occurs, packet loss concealment technology can be used to recover audio data and improve audio call quality. The decoding end builds a model based on the linear prediction coefficient (Linear Prediction Coefficient, LPC) of the previous voice frame, reconstructs the voice signal based on historical voice information, and then loads a certain amount of random noise. When continuous packet loss is concealed, the same linear prediction coefficient LPC is used to reconstruct the voice signal. At the same time, in order to reduce the correlation between consecutive reconstructed signals, the energy of the data packet generated by packet loss concealment decreases; finally, for voice continuity, inter-frame smoothing is also required. When packet loss compensation is required, the latest frame of data is taken out from the voice buffer stored in the last 100ms, and the LPC coefficient of the frame is calculated to replace the current lost frame to achieve the purpose of packet loss compensation. Such as Figure 4 As shown, Figure 4 The upper figure in the figure is the original packet loss audio signal. Figure 4 The lower figure shows the audio signal after packet loss compensation by PLC algorithm. Figure 4 It can be seen that the lost audio signal is effectively compensated.

[0064] In some embodiments, calculating the audio quick play data and the audio stretching data specifically includes:

[0065] Decode the audio data packet to obtain an audio PCM signal;

[0066] Combined with the voice control speed change factor, determine whether the input audio PCM signal needs speed change processing;

[0067] When it is determined that the audio PCM signal needs speed change processing, short-time energy estimation and similarity test are performed on the audio frames of the audio PCM signal, and waveform superposition is performed through the quality control factor to obtain an output voice signal;

[0068] When it is determined that the audio PCM signal does not need speed change processing, the input audio PCM signal is used as the output voice signal.

[0069] Specifically, the audio fast play algorithm and the audio stretching algorithm are both audio speed change algorithms. They shrink or stretch the audio on the time axis while ensuring that the audio does not change its pitch. They use the waveform similarity superposition method. The basic process is as follows: Figure 5 As shown, is the quality control factor, The minimum value of the quality control factor.

[0070] like Figure 5 As shown, calculating the audio quick play data and the audio stretching data includes the following steps:

[0071] S1: decoded audio PCM signal First, combine the voice control speed change factor to determine whether speed change processing is required. If speed change processing is not required, the original voice signal is directly output. ,at this time = ;

[0072] S2: Perform short-time energy estimation on the audio frame that needs speed change processing. The calculation method of the short-time energy estimation of the input signal is as follows:

[0073]

[0074] in, For short-term energy, is the length of the audio frame, is the input audio frame signal; in particular, if the audio frame energy is less than -50db (decibel), the human ear cannot distinguish it, and it is considered a silent frame, and the silent frame is processed separately;

[0075] S3: For the speech frame, similarity measurement is performed by calculating the autocorrelation function. The calculation method is as follows:

[0076]

[0077] in, is the calculated autocorrelation value, is the input speech signal, N is the length of the signal, and k is the time delay;

[0078] S4: The waveform is superimposed by the quality control factor to realize the fast play and stretching of the audio. The waveform superposition calculation method is as follows (using the Hamming window for superposition):

[0079]

[0080] in, To output the speech signal, is the speech signal to be processed, is the Hamming window, s is the frame shift, and L is the signal length.

[0081] In some embodiments, the audio freeze includes major freeze and minor freeze; when the evaluation parameter satisfies a preset evaluation condition, determining that the target audio has audio freeze at the current moment specifically includes:

[0082] When the evaluation parameter satisfies a preset first evaluation condition, determining that a major freeze occurs in the target audio at the current moment;

[0083] When the evaluation parameter satisfies a preset second evaluation condition, it is determined that a slight freeze occurs in the target audio at the current moment.

[0084] The first evaluation condition includes at least one of the following:

[0085] Within the preset time period, the total number of active packet losses and passive packet losses exceeds the first preset time period (e.g. 100ms);

[0086] Within two consecutive preset time periods (e.g., 200ms), the number of active packet losses and passive packet losses within each preset time period exceeds the second preset time period (e.g., 80ms);

[0087] The number of packet loss compensations performed continuously exceeds the preset compensation times within the preset time (for example, 200ms).

[0088] Within a preset time (e.g. 200ms), the data lost during fast playback exceeds the first preset time;

[0089] Within a preset time length (e.g., 200 ms), the data expanded by stretching exceeds a third preset time length;

[0090] Within a preset time (e.g. 200ms), the packet loss compensation process exceeds the preset number of compensation processes;

[0091] Within a preset time length (for example, 200 ms), the average audio jitter value is greater than a fourth preset time length.

[0092] In a specific usage scenario, in the audio signal quality assessment calculation, firstly, according to the auditory characteristics of the human ear, the subdivision unit length of the quality assessment is defined as 200 milliseconds, and then the subdivision unit length of 200 milliseconds is processed; two audio jam values ​​are defined as the quality standard for determining the subdivision unit length, named large jam and small jam respectively, and then the following situations are processed separately:

[0093] The criteria for determining a major jam is the first evaluation condition mentioned above:

[0094] a. The total number of active and passive packet losses within 200ms exceeds 100ms;

[0095] b. In two consecutive 200ms, if the number of active and passive packet losses in each 200ms exceeds 80ms, it is considered that there is a major jam in these 400ms;

[0096] c. Packet loss compensation is performed more than 3 times in a row within 200ms;

[0097] d. The data lost during fast playback exceeds 100ms within 200ms;

[0098] e. The data expanded by stretching within 200ms exceeds 70ms;

[0099] f. Packet loss compensation is processed more than 5 times within 200ms;

[0100] g. The average audio jitter value within 200ms is greater than 500ms;

[0101] The criteria for determining a minor lag are the second evaluation criteria above:

[0102] a. The total amount of active and passive packet losses within 200ms is greater than 20ms and less than or equal to 80ms;

[0103] b. The data lost in fast playback exceeds 60ms within 200ms;

[0104] c. The data expanded by stretching within 200ms exceeds 45ms;

[0105] d. Packet loss compensation is processed more than 2 times and less than 4 times within 200ms;

[0106] e. The average audio jitter value within 200ms is greater than 200ms and less than 500ms.

[0107] After the quality standard of the 200-ms subdivision unit length is determined, the reporting period reported to the background is defined as 4 seconds, so it is also necessary to count the number of major and minor freezes within 4 seconds, and then convert them into scores to express them in the form of scores so that the results can be seen more intuitively; here, the weight of major freezes is defined as 0.7, and the weight of minor freezes is defined as 0.3, and then the freeze value is normalized to a score range of 0 to 5 points, so that the current audio quality can be observed in real time and intuitively in the background.

[0108] In the above specific implementation, the audio signal quality evaluation method provided by the present invention obtains the audio packet data of the target audio to be evaluated, calculates the evaluation parameters of the audio packet data; when the evaluation parameters meet the preset evaluation conditions, determines that the target audio has audio freeze at the current moment; calculates the number of audio freezes within the evaluation time; and generates the evaluation score of the target audio according to the mapping relationship between the preset number of freezes and the evaluation score. After receiving the audio packet information, audio packet loss calculation, audio jitter value calculation, PLC packet loss compensation calculation, audio quick broadcast information calculation and audio stretching information calculation are performed, and then the final result is obtained according to the calculation results and the preset strategy. This method can display the audio signal quality status of all time points in the current call process in the form of scores in real time in the background. Compared with the need to refer to multiple parameters to evaluate the current audio signal quality, it is more intuitive, concise, convenient, and improves accuracy. The device can display the audio signal quality status of all time points in the current call process in the form of scores in the background in real time. Compared with the need to refer to multiple parameters to evaluate the current audio signal quality, it is more intuitive, concise, convenient, and has improved accuracy. At the same time, the device does not need to store additional audio files for comparison, and can obtain the current audio signal quality status, which saves storage space and processing time compared to the audio signal quality evaluation method with reference. The current audio signal quality status can be obtained without human participation, which saves a lot of manpower and time costs compared to manual listening, thereby realizing reference-free evaluation of audio signal quality and improving the convenience, accuracy and real-time performance of audio signal quality evaluation.

[0109] In addition to the above method, the present invention also provides an audio signal quality assessment device, such as Figure 6 As shown, the device comprises:

[0110] The data acquisition unit 610 is used to acquire the audio packet data of the target audio to be evaluated;

[0111] A parameter calculation unit 620, configured to calculate evaluation parameters of the audio packet data;

[0112] A jam calculation unit 630 is used to determine that the target audio has audio jamming at the current moment when the evaluation parameter meets the preset evaluation condition, and calculate the number of audio jamming occurring within the evaluation time;

[0113] The result generating unit 640 is used to generate the evaluation score of the target audio according to the mapping relationship between the preset number of freeze times and the evaluation score.

[0114] In some embodiments, the evaluation parameters of the audio packet data include an audio packet loss rate, an audio jitter value, a packet loss compensation value, audio quick play data, and audio stretching data.

[0115] In some embodiments, the audio jitter value is calculated using a first expression , the first expression is:

[0116]

[0117] in, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet, Representative The sending timestamp of each data packet, Representative The receiving timestamp of the data packet.

[0118] In some embodiments, calculating the audio quick play data and the audio stretching data specifically includes:

[0119] Decode the audio data packet to obtain an audio PCM signal;

[0120] Combined with the voice control speed change factor, determine whether the input audio PCM signal needs speed change processing;

[0121] When it is determined that the audio PCM signal needs speed change processing, short-time energy estimation and similarity test are performed on the audio frames of the audio PCM signal, and waveform superposition is performed through the quality control factor to obtain an output voice signal;

[0122] When it is determined that the audio PCM signal does not need speed change processing, the input audio PCM signal is used as the output voice signal.

[0123] In some embodiments, the audio freeze includes major freeze and minor freeze; when the evaluation parameter satisfies a preset evaluation condition, determining that the target audio has audio freeze at the current moment specifically includes:

[0124] When the evaluation parameter satisfies a preset first evaluation condition, determining that a major freeze occurs in the target audio at the current moment;

[0125] When the evaluation parameter satisfies a preset second evaluation condition, it is determined that a slight freeze occurs in the target audio at the current moment.

[0126] In some embodiments, the first evaluation condition includes at least one of the following:

[0127] Within the preset time period, the total number of active packet losses and passive packet losses exceeds the first preset time period;

[0128] In two consecutive preset time periods, the number of active packet losses and passive packet losses in each preset time period exceeds the second preset time period;

[0129] Within the preset time, the number of consecutive packet loss compensations exceeds the preset compensation times.

[0130] Within the preset time, the data lost by fast playback exceeds the first preset time;

[0131] Within the preset time period, the data expanded by stretching exceeds a third preset time period;

[0132] Within the preset time, the packet loss compensation process exceeds the preset compensation process times;

[0133] Within the preset time length, the average audio jitter value is greater than a fourth preset time length.

[0134] In the above specific implementation, the audio signal quality assessment device provided by the present invention obtains the audio packet data of the target audio to be assessed, calculates the assessment parameters of the audio packet data; when the assessment parameters meet the preset assessment conditions, determines whether the target audio has audio freeze at the current moment; calculates the number of audio freezes within the assessment time; and generates the assessment score of the target audio based on the mapping relationship between the preset number of freezes and the assessment score. After receiving the audio packet information, audio packet loss calculation, audio jitter value calculation, PLC packet loss compensation calculation, audio quick broadcast information calculation and audio stretching information calculation are performed, and then the final result is obtained based on the calculation results and the preset strategy. The device can display the audio signal quality status of all time points in the current call process in the form of scores in the background in real time. Compared with the need to refer to multiple parameters to evaluate the current audio signal quality, it is more intuitive, concise, convenient, and has improved accuracy. At the same time, the device does not need to store additional audio files for comparison, and can obtain the current audio signal quality status, which saves storage space and processing time compared to the audio signal quality evaluation method with reference. The current audio signal quality status can be obtained without human participation, which saves a lot of manpower and time costs compared to manual listening, thereby realizing reference-free evaluation of audio signal quality and improving the convenience, accuracy and real-time performance of audio signal quality evaluation.

[0135] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the above method.

[0136] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0137] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the above method.

[0138] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0139] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating audio signal quality, It is characterized in that The method comprises: Obtain audio packet data of the target audio to be evaluated; Calculating evaluation parameters of the audio packet data; When the evaluation parameter satisfies a preset evaluation condition, determining that audio freeze occurs in the target audio at the current moment; Calculate the number of audio freezes that occur during the evaluation period; Generate an evaluation score for the target audio according to a preset mapping relationship between the number of freezes and the evaluation score; The evaluation parameters of the audio packet data include audio packet loss rate, audio jitter value, packet loss compensation value, audio fast play data and audio stretching data; Calculate the audio jitter value using the first expression , the first expression is: in, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet, Representative The sending timestamp of each data packet, Representative The receiving timestamp of each data packet; Calculate audio fast play data and audio stretch data, including: Decode the audio data packet to obtain an audio PCM signal; Combined with the voice control speed change factor, determine whether the input audio PCM signal needs speed change processing; When it is determined that the audio PCM signal needs speed change processing, short-time energy estimation and similarity test are performed on the audio frames of the audio PCM signal, and waveform superposition is performed through the quality control factor to obtain an output voice signal; When it is determined that the audio PCM signal does not need speed change processing, the input audio PCM signal is used as the output voice signal.

2. The audio signal quality assessment method according to claim 1, It is characterized in that The audio freeze includes major freeze and minor freeze; when the evaluation parameter satisfies a preset evaluation condition, determining that the target audio has audio freeze at the current moment specifically includes: When the evaluation parameter satisfies a preset first evaluation condition, determining that a major freeze occurs in the target audio at the current moment; When the evaluation parameter satisfies a preset second evaluation condition, it is determined that a slight freeze occurs in the target audio at the current moment.

3. The audio signal quality assessment method according to claim 2, It is characterized in that The first evaluation condition includes at least one of the following: Within the preset time period, the total number of active packet losses and passive packet losses exceeds the first preset time period; In two consecutive preset time periods, the number of active packet losses and passive packet losses in each preset time period exceeds the second preset time period; Within the preset time, the number of consecutive packet loss compensations exceeds the preset compensation times; Within the preset time, the data lost by fast playback exceeds the first preset time; Within the preset time period, the data expanded by stretching exceeds a third preset time period; Within the preset time, the packet loss compensation process exceeds the preset compensation process times; Within the preset time length, the average audio jitter value is greater than a fourth preset time length.

4. An audio signal quality assessment device, used to implement the method according to any one of claims 1 to 3, It is characterized in that The device comprises: A data acquisition unit, used to acquire audio packet data of the target audio to be evaluated; a parameter calculation unit, configured to calculate evaluation parameters of the audio packet data; A jam calculation unit, configured to determine that audio jam occurs in the target audio at a current moment when the evaluation parameter satisfies a preset evaluation condition, and calculate the number of times the audio jam occurs within the evaluation time; A result generating unit is used to generate an evaluation score of the target audio according to a preset mapping relationship between the number of freezes and the evaluation score.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the program, the method according to any one of claims 1 to 3 is implemented.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

7. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.