Communication voice quality evaluation method and device and electronic equipment

Abstract

The invention discloses a communication voice quality evaluation method and device and electronic equipment. The method comprises the steps that a signal to be processed is acquired, the signal to be processed is divided into a target time frame with a preset length, and the preset length ranges from 50 milliseconds to 100 milliseconds; splitting the target time frame into a plurality of sub-blocks, and determining the Bark scale of each sub-block according to the scaling factor and the frequency of the target time frame; fusing the Bark scales of the plurality of sub-blocks to obtain a target Bark scale; and determining the target loudness of the target time frame according to the target Bark scale. According to the invention, the technical problem of inaccurate voice quality score caused by frame boundary mismatch and insufficient time domain masking effect when voice quality evaluation is carried out by using a traditional PESQ algorithm with short-time frame length due to relatively large signal propagation time delay in high-orbit satellite communication is solved.

Description

Technical Field

[0001] This application relates to the field of voice quality assessment, and more specifically, to a method, apparatus, and electronic device for assessing the quality of communication voice. Background Technology

[0002] In modern satellite communication systems, Geostationary Earth Orbit (GEO) satellites are widely used to provide global voice communication services due to their fixed location advantage. However, because GEO satellites are approximately 36,000 kilometers above the Earth's surface, the one-way propagation delay of the signal is about 120 ms, and the round-trip time (RTT) is about 250 ms. Compared to terrestrial mobile communication networks (such as 4G and 5G), whose signal propagation delay is typically between a few milliseconds and tens of milliseconds, GEO satellite communication faces significant latency challenges.

[0003] The Perceptual Evaluation of Speech Quality (PESQ) algorithm, defined by the ITU-T P.862 standard, is one of the most commonly used objective speech quality assessment tools. The PESQ algorithm primarily quantifies speech quality by comparing the energy distribution of a reference speech signal and a damaged speech signal across different frequency bands, and by taking into account the masking effect of the human auditory system. To match the perceptual characteristics of human ears, the PESQ algorithm uses a 32ms frame length with 50% overlap between frames to reduce spectral leakage, improve temporal continuity, and avoid frame boundary distortion.

[0004] However, in high-orbit satellite communication scenarios, a frame length of 32ms may not accurately capture the changes of the entire speech unit, especially with longer latency. The preceding and following frames of the speech signal may no longer conform to the expected temporal relationship, causing speech units at frame boundaries to be cut off or overlapped, thus affecting the computational accuracy of the PESQ algorithm. Furthermore, due to signal transmission delay, the synchronization problem of the speech signal becomes more prominent, and traditional frame length settings may lead to difficulties in signal alignment, affecting the evaluation results.

[0005] Furthermore, the temporal masking effect evaluation in the PESQ algorithm assumes a relatively short masking window, approximately 50ms. This is reasonable in terrestrial networks because the signal delay is small, allowing the masking effect to be fully manifested within a single frame. However, in high-orbit satellite communications, the additional signal delay may prevent the masking effect from being fully reflected in the current frame, or leave lingering masking effects at the start of the next frame, which can also lead to evaluation distortion.

[0006] In summary, the large latency characteristic of high-orbit satellite communication conflicts with the inherent short frame length and temporal masking window setting of the PESQ algorithm, resulting in insufficient accuracy of the PESQ algorithm in evaluating the quality of high-orbit satellite voice communication, and thus requiring optimization.

[0007] There is currently no effective solution to the above problems. Summary of the Invention

[0008] This application provides a communication voice quality assessment method, apparatus, and electronic device to at least solve the technical problem of inaccurate voice quality scoring caused by frame boundary mismatch and insufficient temporal masking effect when using the traditional short-time frame-length PESQ algorithm for voice quality assessment due to the large signal propagation delay in high-orbit satellite communication.

[0009] According to one aspect of this application, a method for evaluating the quality of communication voice is provided, comprising: acquiring a signal to be processed; dividing the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; dividing the target time frames into multiple sub-blocks, and determining the Bark scale of each sub-block according to a scaling factor and the frequency of the target time frame; fusing the Bark scales of the multiple sub-blocks to obtain a target Bark scale; and determining the target loudness of the target time frame according to the target Bark scale.

[0010] Optionally, the Bark scale of each sub-block can be determined using the following formula. : Where f is the frequency of the signal to be processed. Scaling factor The value range is from 0.8 to 1.2.

[0011] Optionally, the target time frame can be divided into multiple sub-blocks, including: dividing the target time frame into two sub-blocks when the preset length is 50 milliseconds; and dividing the target time frame into four sub-blocks when the preset length is 100 milliseconds.

[0012] Optionally, after determining the Bark scale of each sub-block, the method further includes: if the frequency of the signal to be processed is within a first range when the preset length is 50 milliseconds, dividing the Bark scale frequency band corresponding to the first range into eight groups, wherein the first range includes: less than 1 kilohertz; if the frequency of the signal to be processed is within the first range when the preset length is 100 milliseconds, dividing the Bark scale frequency band corresponding to the first range into twelve groups; if the frequency of the signal to be processed is within a second range when the preset length is 50 milliseconds, merging the Bark scale frequency band corresponding to the second range into six groups, wherein the second range includes: greater than 4 kilohertz; if the frequency of the signal to be processed is within the second range when the preset length is 100 milliseconds, merging the Bark scale frequency band corresponding to the second range into four groups.

[0013] Optionally, the target loudness of the target time frame can be determined using the following formula. : Where L is the sound pressure level of the target time frame.

[0014] Optionally, the method further includes: calculating the masking threshold using the following formula. : ,in, For the loudness of the covert, This refers to the frequency band distance.

[0015] Optionally, the target loudness of the target time frame can be determined using the following formula. : .

[0016] According to another aspect of this application, a communication voice quality assessment device is also provided, comprising: an acquisition module for acquiring a signal to be processed and dividing the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; a first determination module for dividing the target time frames into multiple sub-blocks and determining the Bark scale of each sub-block according to a scaling factor and the frequency of the target time frames; a fusion module for fusing the Bark scales of the multiple sub-blocks to obtain a target Bark scale; and a second determination module for determining the target loudness of the target time frames according to the target Bark scale.

[0017] According to another aspect of this application, a non-volatile storage medium is also provided, the storage medium including a stored program, wherein the program, when running, controls the device where the storage medium is located to execute the above-mentioned communication voice quality evaluation method.

[0018] According to another aspect of this application, an electronic device is also provided, comprising: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes the above-described communication voice quality evaluation method during runtime.

[0019] According to another aspect of this application, a computer program is also provided, wherein the computer program, when executed by a processor, implements the above-described communication voice quality assessment method.

[0020] According to another aspect of this application, a computer program product is also provided, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-described communication voice quality evaluation method.

[0021] In this application, the following methods are employed: acquiring the signal to be processed, dividing the signal into target time frames of a preset length (50 to 100 milliseconds); splitting the target time frames into multiple sub-blocks, and determining the Bark scale of each sub-block based on a scaling factor and the frequency of the target time frame; fusing the Bark scales of multiple sub-blocks to obtain the target Bark scale; and determining the target loudness of the target time frame based on the target Bark scale. By dynamically adjusting the frame length and fine-grained frequency grouping, this approach adapts to the large latency characteristics of high-orbit satellite communication and optimizes the accuracy of speech quality assessment. This achieves accurate matching of speech unit changes even with long signal propagation delays in a high-orbit satellite environment, effectively improving the assessment of temporal masking effects. Furthermore, it solves the technical problem of inaccurate speech quality scoring caused by frame boundary mismatch and insufficient temporal masking effects when using traditional short-frame-length PESQ algorithms for speech quality assessment due to the large signal propagation delays in high-orbit satellite communication. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0023] Figure 1 This is a flowchart of a communication voice quality assessment method according to an embodiment of this application;

[0024] Figure 2 This is a structural diagram of a communication voice quality assessment device according to an embodiment of this application;

[0025] Figure 3 This is a hardware structure block diagram of a computer terminal for a communication voice quality assessment method according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] According to an embodiment of this application, a method embodiment for evaluating the quality of communication voice is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] Figure 1 This is a flowchart of a communication voice quality assessment method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0030] Step S102: Obtain the signal to be processed and divide the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds.

[0031] In step S102, the signal to be processed, such as a voice signal, is first acquired. The voice signal originates from any voice communication process within the high-orbit satellite communication system. The acquired signal to be processed is divided into a series of target time frames according to a pre-set length. The length of each time frame is set between 50 and 100 milliseconds. This setting directly responds to the characteristic of long signal propagation delay in high-orbit satellite communication scenarios, ensuring that the time frames can cover more complete voice units.

[0032] Step S104: Divide the target time frame into multiple sub-blocks, and determine the Bark scale of each sub-block based on the scaling factor and the frequency of the target time frame.

[0033] In step S104, each target time frame is further subdivided into several sub-blocks, the number of which depends on the specific length of the time frame. For example, a 50-millisecond time frame may be divided into 2 sub-blocks, while a 100-millisecond time frame may be divided into 4 sub-blocks. Then, a specific scaling factor is applied to the frequency analysis of each sub-block to adjust the calculation of the Bark scale. The scaling factor ranges from 0.8 to 1.2, dynamically compressing or expanding the high-frequency bandwidth to ensure that frequency grouping considers both the frequency perception characteristics of the human ear and the needs of increased frame length. Through this method, the Bark scale is calculated independently for each sub-block, achieving optimized processing of the frequency nonlinear mapping.

[0034] Step S106: Fuse the Bark scales of multiple sub-blocks to obtain the target Bark scale.

[0035] In step S106, the Bark scales of the multiple sub-blocks obtained in step S104 are fused to generate a unified Bark scale for the entire target time frame. The fusion process aims to balance the frequency characteristics between different sub-blocks, ensuring that the frequency analysis within the entire frame is both detailed and comprehensive, reflecting the overall frequency energy distribution. Fusion strategies include, but are not limited to, weighted averaging or statistical synthesis. The key is to ensure that the final Bark scale can represent the frequency characteristics of the complete time frame, while also taking into account the special time delay conditions in high-orbit satellite communication.

[0036] Step S108: Determine the target loudness of the target time frame based on the target Bark scale.

[0037] Step S108 includes determining the target loudness of the target time frame based on the fused target Bark scale. The target loudness refers to the loudness value of the speech signal within the time frame, calculated considering the non-linear perception of volume by the human ear. Specifically, the power of all frequency points within the time frame is converted to obtain a loudness value expressed in Son units. Furthermore, the temporal masking effect is specifically considered; by adjusting the masking threshold, the loudness calculation process accurately reflects the masking effect of strong signals on neighboring weak signals, resulting in a target loudness value that is closer to actual human perception, thus providing a more objective speech quality score.

[0038] In summary, the above steps achieve a more accurate and adaptable voice quality assessment of high-orbit satellite voice communication signals. Specifically, after acquiring the voice signal to be processed in step S102, it is divided into target time frames of a preset length, set between 50 and 100 milliseconds. This ensures that each time frame can more comprehensively capture the essential characteristics of the voice signal, especially the integrity and temporal continuity of voice units, given the significantly increased signal propagation delay in high-orbit satellite communication. This reduces assessment errors caused by mismatches between frame boundaries and the natural boundaries of the voice signal. Step S104 further divides each target time frame into multiple sub-blocks, effectively avoiding the problem of reduced high-frequency resolution due to increased frame length. By introducing a dynamic scaling factor to adjust the Bark scale, the Bark scale is calculated independently for each sub-block based on the frequency information in the target time frame. This not only maintains the original frequency resolution capability but also further optimizes frequency grouping, particularly improving the accurate processing of high-frequency signals, ensuring that even under large time delay conditions, the frequency perception characteristics of the human ear can still be accurately reflected. In step S106, the Bark scales of multiple sub-blocks are fused to generate the target Bark scale. The fusion process comprehensively considers the frequency characteristics of different sub-blocks, effectively avoiding the adverse effects of local characteristics on the overall evaluation results through weighting or other appropriate methods, ensuring the consistency and accuracy of frequency energy distribution within the target time frame. Step S108 determines the target loudness of the target time frame based on the obtained target Bark scale. Through a redesigned Sone unit calculation formula, the power spectrum under the Bark scale is converted into a loudness value. Simultaneously, considering the correction of temporal masking effects, the loudness calculation better reflects the true hearing experience of the human ear. This can greatly improve the objectivity and reliability of speech quality evaluation in high-orbit satellite voice communication scenarios, avoiding the impact of signal propagation delay and ensuring a high degree of consistency between the evaluation results and human subjective perception.

[0039] In summary, this application successfully solved the technical challenges of using the traditional PESQ algorithm for voice quality assessment in high-orbit satellite communications by increasing frame length, refining frequency grouping, and optimizing loudness calculation. This significantly improved the assessment accuracy and provided strong technical support for optimizing the quality of high-orbit satellite voice communication services.

[0040] The following are Figure 1 The steps shown are illustrated and explained by way of example.

[0041] According to some optional embodiments of this application, the Bark scale of each sub-block is determined by the following formula. : Where f is the frequency of the signal to be processed. Scaling factor The value range is from 0.8 to 1.2.

[0042] The scaling factor (k) can dynamically adjust the width of the Bark band, especially at long frame times (50 to 100 milliseconds), effectively avoiding the problem of excessively coarse frequency resolution in the high-frequency band. By appropriately compressing or expanding the high-frequency bandwidth, not only can sensitivity to low-frequency signals be maintained, but the accuracy of high-frequency signal details can also be preserved. This is particularly important in high-orbit satellite communication scenarios, because information loss in high-frequency signals is more difficult to detect and compensate for than in low-frequency signals.

[0043] Furthermore, signal characteristics (such as signal-to-noise ratio and interference type) vary in different communication scenarios. The flexibility of the scaling factor (k) allows the algorithm to better adapt to these changes, especially the significant time delay and potential signal attenuation present in high-orbit satellite communications. By fine-tuning the Bark scale calculation, the signal frequency distribution under various communication environments can be reflected more accurately, improving the accuracy and stability of voice quality assessment.

[0044] Furthermore, the scaling factor (k) can enhance the robustness of the PESQ algorithm, making it more resilient in handling various complex signals. In high-orbit satellite communications, signals may experience multiple reflections, multipath propagation, and long propagation delays, all of which can affect the signal's frequency characteristics. By appropriately adjusting the value of (k), accurate assessment of signal quality can be maintained even in the face of these uncertainties, avoiding the assessment biases inherent in traditional fixed parameter settings.

[0045] Furthermore, it's important to note that the Bark scale was designed to simulate the human ear's perception of different frequencies. The addition of a scaling factor (k) makes this simulation more realistic. While the human ear has relatively low resolution in the high-frequency range, long frame times can lead to oversimplification in high-frequency analysis. By dynamically adjusting the Bark scale, the perceptual capabilities of the human ear in the high-frequency range can be more accurately reflected, ensuring that speech quality assessment results better align with human auditory experience.

[0046] According to some other optional embodiments of this application, the target time frame can be divided into multiple sub-blocks by the following methods: when the preset length is 50 milliseconds, the target time frame is divided into two sub-blocks; when the preset length is 100 milliseconds, the target time frame is divided into four sub-blocks.

[0047] It should be noted that in high-orbit satellite communication, signal propagation delay increases significantly, and traditional short-frame-length algorithms may not be fully suitable. By splitting the signal into sub-blocks, analyzing each block independently, and then fusing them, longer signal segments can be processed without sacrificing temporal resolution. This is crucial for matching the signal propagation characteristics of high-orbit satellite communication, as it maintains good evaluation performance even under large delay conditions. The process of splitting and fusing sub-blocks increases the computational burden of the algorithm to some extent, but while improving accuracy, the efficiency loss caused by excessive subdivision is avoided through reasonable control of the number of sub-blocks (e.g., splitting 50 milliseconds into 2 sub-blocks, and 100 milliseconds into 4 sub-blocks). This balancing strategy ensures that the algorithm can achieve the expected evaluation accuracy while maintaining high computational efficiency when processing high-orbit satellite speech signals, making it suitable for real-time or near-real-time speech communication quality monitoring.

[0048] Dividing the target time frame into sub-blocks helps to capture the frequency characteristics of the signal more accurately. Although the 50-millisecond frame length is longer than the 32-millisecond frame length of the traditional PESQ algorithm, it still remains within a relatively compact time range. By dividing it into two sub-blocks, each 25 milliseconds long, we can not only take advantage of the longer frame to reduce spectral leakage, but also maintain the frequency resolution within a shorter time interval, which is beneficial for detailed analysis of rapidly changing components in the speech signal.

[0049] With a frame length of 100 milliseconds, the target time frame is further divided into four sub-blocks, each with a length of 25 milliseconds. This helps to better preserve the details of high-frequency signals. High-frequency signals typically change rapidly and play a crucial role in human auditory perception, especially in speech intelligibility and sound quality. Sub-block processing ensures that the algorithm can independently focus on the frequency information of each sub-block, thus avoiding the high-frequency resolution degradation caused by long frame lengths and making the evaluation of high-frequency components more accurate.

[0050] Optionally, as the preset length increases, the number of sub-blocks should increase accordingly to maintain the computational efficiency and accuracy of the algorithm. For example, if the preset length is 75 milliseconds, it can be divided into three sub-blocks, each 25 milliseconds long. This can reduce spectral leakage and improve temporal continuity by utilizing the longer frame time, while avoiding excessive computational load.

[0051] In addition, a baseline length (e.g., 25 milliseconds) can be set, and the number of sub-blocks can be dynamically determined based on the ratio of the preset length to the baseline length. If the preset length is a multiple of the baseline length, the number of sub-blocks is directly set to the result of dividing the preset length by the baseline length; if it is not an integer multiple, the closest and most reasonable number of sub-blocks can be determined through rounding or other logical rules. This dynamic adjustment ensures that the algorithm can adapt to various frame lengths while maintaining an optimized level of signal processing.

[0052] In some optional embodiments of this application, after determining the Bark scale of each sub-block, the following steps can also be performed: If the frequency of the signal to be processed is within a first range when the preset length is 50 milliseconds, the Bark scale frequency band corresponding to the first range is divided into eight groups, wherein the first range includes: less than 1 kilohertz; If the frequency of the signal to be processed is within the first range when the preset length is 100 milliseconds, the Bark scale frequency band corresponding to the first range is divided into twelve groups; If the frequency of the signal to be processed is within a second range when the preset length is 50 milliseconds, the Bark scale frequency band corresponding to the second range is merged into six groups, wherein the second range includes: greater than 4 kilohertz; If the frequency of the signal to be processed is within the second range when the preset length is 100 milliseconds, the Bark scale frequency band corresponding to the second range is merged into four groups.

[0053] It's worth explaining that, with a preset length of 50 milliseconds, if the signal frequency is in the first range below 1 kHz, this frequency band is further subdivided into eight groups. Similarly, at 100 milliseconds, the first low-frequency range is further divided into twelve groups. This division method enhances the ability to capture details in the low-frequency band. Low-frequency information is crucial for speech clarity and comprehension. By increasing the number of groups, the characteristics of low-frequency sounds, such as fundamental frequency and formant positions, can be more finely distinguished and analyzed, playing an important role in improving the accuracy of speech quality assessment.

[0054] For the second frequency range above 4 kHz, the high-frequency Bark bands are merged into six groups with a preset length of 50 milliseconds; and further merged into four groups with a preset length of 100 milliseconds. Although the high-frequency band contains rich detail, the resolution in human auditory perception decreases as the frequency increases. By merging the bands, unnecessary computation can be reduced, improving processing efficiency, while also ensuring that the algorithm can focus on the high-frequency components that are truly meaningful for speech quality assessment, effectively balancing the relationship between accuracy and efficiency.

[0055] The increase in the preset frame length (from 50 milliseconds to 100 milliseconds) is to better accommodate the high latency characteristics of high-orbit satellite communication. However, longer frame times may lead to an increase in the number of FFT points, and improper frequency band grouping will affect computation speed and resource consumption. Through the aforementioned frequency band grouping strategy, good response speed can still be maintained when processing long frame data. At the same time, refined low-frequency processing and simplified high-frequency processing ensure the algorithm's comprehensiveness and relevance in analyzing signal frequency characteristics, making the evaluation results more closely reflect the realities of high-orbit satellite communication.

[0056] As some optional embodiments of this application, the target loudness of the target time frame is determined by the following formula. : Where L is the sound pressure level of the target time frame.

[0057] This embodiment is in the high-loudness region. Increasing the coefficient in the traditional formula from 0.5 to 0.6 means that the loudness assessment of loud signals is closer to the actual human ear's perception. This is because the human ear's response to high-loudness signals does not increase linearly, but gradually levels off. A higher coefficient can reflect this non-linear response characteristic, thus making the loudness value output by the algorithm closer to human subjective hearing, improving the accuracy of the assessment. In the low-loudness region... Increasing the coefficient from 0.25 to 0.3 indicates that the algorithm becomes more sensitive to low-loudness signals. In communication environments, low-loudness signals are often more susceptible to factors such as noise and transmission loss, while the human ear is more sensitive to low-loudness signals. Therefore, increasing the coefficient helps the algorithm more accurately reflect the loudness level and its quality changes in the communication link when processing such signals.

[0058] In real-world communication scenarios, especially in high-orbit satellite communication, signals may experience significant attenuation and noise interference, leading to large fluctuations in sound pressure levels. By adjusting the coefficients, these variability can be better accommodated, providing a more reasonable loudness assessment that closely reflects real-world experience, even under conditions of poor signal-to-noise ratio.

[0059] In some optional embodiments of this application, the communication voice quality assessment method further includes the following step: calculating a masking threshold using the following formula. : ,in, For the loudness of the covert, This refers to the frequency band distance.

[0060] Furthermore, the target loudness of the target time frame can be determined using the following formula. : .

[0061] In this embodiment, the masking threshold The calculation adjusted the bandwidth distance using a coefficient of 0.15. The degree of impact on the masking effect. In high-orbit satellite communication scenarios, the frequency components of the signal may change, and the mutual masking effect between different frequencies will also change accordingly. By adjusting the masking threshold, these changes can be reflected more accurately, improving the objectivity of the assessment.

[0062] Target loudness The calculation is actually an assessment of the loudness of the damaged signal. An adjustment based on the masking effect was made, which allows the algorithm to more finely evaluate the true quality of the speech signal, especially in scenarios where strong signals mask weak signals. This is achieved by subtracting a masking threshold. It can more accurately capture those signal details that are easily overlooked, which is especially crucial for scenarios like high-orbit satellite communication where signal integrity is highly critical.

[0063] Figure 2 This is a structural diagram of a communication voice quality assessment device according to an embodiment of this application, such as... Figure 2 As shown, the device includes:

[0064] The acquisition module 22 is used to acquire the signal to be processed and divide the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds.

[0065] The first determining module 24 is used to divide the target time frame into multiple sub-blocks and determine the Bark scale of each sub-block according to the scaling factor and the frequency of the target time frame.

[0066] The fusion module 26 is used to fuse the Bark scale of multiple sub-blocks to obtain the target Bark scale.

[0067] The second determining module 28 is used to determine the target loudness of the target time frame based on the target Bark scale.

[0068] Optionally, the Bark scale of each sub-block can be determined using the following formula. : Where f is the frequency of the signal to be processed. Scaling factor The value range is from 0.8 to 1.2.

[0069] Optionally, the target time frame can be divided into multiple sub-blocks, specifically including the following steps: when the preset length is 50 milliseconds, the target time frame can be divided into two sub-blocks; when the preset length is 100 milliseconds, the target time frame can be divided into four sub-blocks.

[0070] Optionally, after determining the Bark scale of each sub-block, the following steps can also be performed: If the frequency of the signal to be processed is within a first range when the preset length is 50 milliseconds, the Bark scale frequency band corresponding to the first range is divided into eight groups, wherein the first range includes frequencies less than 1 kilohertz; If the frequency of the signal to be processed is within the first range when the preset length is 100 milliseconds, the Bark scale frequency band corresponding to the first range is divided into twelve groups; If the frequency of the signal to be processed is within a second range when the preset length is 50 milliseconds, the Bark scale frequency band corresponding to the second range is merged into six groups, wherein the second range includes frequencies greater than 4 kilohertz; If the frequency of the signal to be processed is within the second range when the preset length is 100 milliseconds, the Bark scale frequency band corresponding to the second range is merged into four groups.

[0071] Optionally, the target loudness of the target time frame can be determined using the following formula. : Where L is the sound pressure level of the target time frame.

[0072] Optionally, the communication voice quality assessment device is also used to perform the following steps: calculating the masking threshold using the following formula. : ,in, For the loudness of the covert, This refers to the frequency band distance.

[0073] Optionally, the target loudness of the target time frame can be determined using the following formula. : .

[0074] It should be noted that the above Figure 2 The modules in can be program modules (e.g., a set of program instructions that implements a specific function) or hardware modules. For the latter, they can be represented in the following forms, but are not limited to these: each of the above modules is represented by a processor, or the functions of each of the above modules are implemented by a processor.

[0075] It should be noted that, Figure 2 Preferred embodiments of the shown examples can be found in [reference needed]. Figure 1 The relevant descriptions of the embodiments shown will not be repeated here.

[0076] Figure 3 A hardware block diagram of a computer terminal for implementing a communication voice quality assessment method is shown. Figure 3As shown, the computer terminal 30 may include one or more processors 302 (shown as 302a, 302b, ..., 302n in the figure) 302 (processor 302 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 304 for storing data, and a transmission module 306 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 3 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 30 may also include... Figure 3 The more or fewer components shown, or having the same Figure 3 The different configurations shown.

[0077] It should be noted that the aforementioned one or more processors 302 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 30. As involved in the embodiments of this application, the data processing circuits serve as processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0078] The memory 304 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the communication voice quality assessment method in this embodiment. The processor 302 executes various functional applications and data processing by running the software programs and modules stored in the memory 304, thereby realizing the aforementioned communication voice quality assessment method. The memory 304 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 304 may further include memory remotely located relative to the processor 302, and these remote memories can be connected to the computer terminal 30 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0079] The transmission module 306 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 30. In one example, the transmission module 306 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 306 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0080] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 30.

[0081] It should be noted here that, in some optional embodiments, the above... Figure 3 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 3 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0082] It should be noted that, Figure 3 The computer terminal shown is used to execute Figure 1 The communication voice quality assessment method shown above applies to this electronic device as well, and will not be repeated here.

[0083] This application also provides a non-volatile storage medium, which includes a stored program, wherein the program, when running, controls the device where the storage medium is located to execute the above-mentioned communication voice quality assessment method.

[0084] A non-volatile storage medium performs the following functions: acquires a signal to be processed; divides the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; divides the target time frames into multiple sub-blocks and determines the Bark scale of each sub-block according to the scaling factor and the frequency of the target time frame; fuses the Bark scales of multiple sub-blocks to obtain the target Bark scale; and determines the target loudness of the target time frame according to the target Bark scale.

[0085] This application also provides an electronic device, including: a memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the program executes the above-described communication voice quality evaluation method when it runs.

[0086] The processor is used to run a program that performs the following functions: acquires the signal to be processed, divides the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; splits the target time frames into multiple sub-blocks, and determines the Bark scale of each sub-block according to the scaling factor and the frequency of the target time frame; fuses the Bark scales of multiple sub-blocks to obtain the target Bark scale; and determines the target loudness of the target time frame according to the target Bark scale.

[0087] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0088] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0089] In the above embodiments of this application, the information collected is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with relevant laws, regulations and standards, take necessary protective measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0091] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0093] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0094] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for evaluating the quality of voice communication, characterized in that, include: Acquire the signal to be processed, and divide the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; The target time frame is divided into multiple sub-blocks, and the Bark scale of each sub-block is determined according to the scaling factor and the frequency of the target time frame. The Bark scales of the multiple sub-blocks are fused to obtain the target Bark scale; The target loudness of the target time frame is determined based on the target Bark scale.

2. The method according to claim 1, characterized in that, The Bark scale for each sub-block is determined using the following formula. : Where f is the frequency of the signal to be processed. The scaling factor, The value range is from 0.8 to 1.

2.

3. The method according to claim 1, characterized in that, The target time frame is divided into multiple sub-blocks, including: With the preset length of 50 milliseconds, the target time frame is split into two sub-blocks; With the preset length of 100 milliseconds, the target time frame is divided into four sub-blocks.

4. The method according to claim 1, characterized in that, After determining the Bark scale for each sub-block, the method further includes: When the preset length is 50 milliseconds, if the frequency of the signal to be processed is in the first range, the Bark scale frequency band corresponding to the first range is divided into eight groups, wherein the first range includes: less than 1 kilohertz; When the preset length is 100 milliseconds, if the frequency of the signal to be processed is within the first range, the Bark scale frequency band corresponding to the first range is divided into twelve groups; When the preset length is 50 milliseconds, if the frequency of the signal to be processed is in the second range, the Bark scale frequency bands corresponding to the second range are merged into six groups, wherein the second range includes: greater than 4 kHz; If the frequency of the signal to be processed is within the second range when the preset length is 100 milliseconds, the Bark scale frequency bands corresponding to the second range are merged into four groups.

5. The method according to claim 1, characterized in that, The target loudness of the target time frame is determined by the following formula. : Where L is the sound pressure level of the target time frame.

6. The method according to claim 5, characterized in that, The method further includes: calculating the masking threshold using the following formula. : ,in, For the loudness of the covert, This refers to the frequency band distance.

7. The method according to claim 6, characterized in that, The target loudness of the target time frame is determined by the following formula. : .

8. A communication voice quality assessment device, characterized in that, include: The acquisition module is used to acquire the signal to be processed and divide the signal to be processed into target time frames of a preset length, wherein the preset length is 50 to 100 milliseconds; The first determining module is used to divide the target time frame into multiple sub-blocks and determine the Bark scale of each sub-block according to the scaling factor and the frequency of the target time frame. The fusion module is used to fuse the Bark scales of the multiple sub-blocks to obtain the target Bark scale; The second determining module is used to determine the target loudness of the target time frame based on the target Bark scale.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the non-volatile storage medium to perform the communication voice quality assessment method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: A memory and a processor, the processor being configured to run a program stored in the memory, wherein the program, when running, executes the communication voice quality assessment method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the communication voice quality assessment method according to any one of claims 1 to 7.

Images