Method for evaluating audio quality and electronic equipment

Abstract

The invention discloses a method for evaluating audio quality and electronic equipment. The method comprises the following steps: acquiring at least two groups of to-be-evaluated parameters of the audio processor; respectively constructing at least two evaluation units corresponding to the at least two groups of to-be-evaluated parameters; respectively combining the at least two evaluation units according to a plurality of configuration strategies to obtain a plurality of evaluation configurations; respectively executing audio processing operation on audio sample data based on the various evaluation configurations so as to generate intermediate files of the at least two evaluation units under the various evaluation configurations; and determining audio quality results of the various evaluation configurations according to differences among the intermediate files of the associated evaluation units in the various evaluation configurations. According to the method, the availability of functions processed by all configurations of the digital audio processor can be covered and evaluated through audio quality evaluation, so that the development efficiency and the test efficiency of the digital audio processor are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital audio processing, and in particular to a method and electronic equipment for evaluating audio quality. Background Art

[0002] With the development of embedded processors and operating systems, users have an increasingly strong demand for high-quality audio, and high-quality audio processor products have emerged in the market to meet users' needs for audio functions such as multi-channel stereo and spatial audio. Nowadays, electronic products such as smart phones, smart cockpits, VR / AR / MR products (i.e. virtual reality technology / augmented reality technology / mixed reality technology products), smart TVs and advanced speakers have very high requirements for high-quality audio.

[0003] The computing power of embedded processors based on ARM architecture is limited in achieving high-quality audio. Currently, embedded processors generally adopt a hybrid architecture, which integrates ARM multi-core processors and DSP audio processors (Digital Signal Processor) to meet the computing needs of high-quality audio.

[0004] During the development of digital audio processors, it is necessary to verify the quality of the audio processed by the internal configuration of the digital audio processor to ensure product quality. However, the digital audio processor has many internal audio interfaces, IPs, and components, and the configuration is complex, so it is difficult to quickly cover and evaluate the functional availability of all configurations. In addition, the output port of the digital audio processor integrates a digital-to-analog converter, which converts the high-quality digital audio processed by the audio algorithm into analog audio, resulting in a large gap between the high-quality audio and the original audio, making it difficult to evaluate the audio quality processed by the digital-to-analog converter. Summary of the invention

[0005] The present invention provides a method and electronic device for evaluating audio quality. The audio quality evaluation can cover and evaluate the availability of functions processed by all configurations of a digital audio processor, so as to improve the development efficiency and test efficiency of the digital audio processor.

[0006] In one aspect of the present invention, a method for evaluating audio quality is provided. The method comprises: obtaining at least two groups of parameters to be evaluated of an audio processor; respectively constructing at least two evaluation units corresponding to the at least two groups of parameters to be evaluated; respectively combining the at least two evaluation units according to multiple configuration strategies to obtain multiple evaluation configurations; respectively performing audio processing operations on audio sample data based on the multiple evaluation configurations to generate intermediate files of the at least two evaluation units under the various evaluation configurations; and determining audio quality results of the various evaluation configurations according to the differences between the intermediate files of the evaluation units associated in the various evaluation configurations.

[0007] In another aspect of the present invention, an electronic device is provided, which includes a memory configured to store a computer program and a processor configured to execute the computer program to perform the above-mentioned method for evaluating audio quality.

[0008] According to the technical solution of the present invention, based on different configuration strategies, the evaluation units are combined to obtain different evaluation configurations, and the configuration is the combination mode of the evaluation units, so as to simulate the multi-configuration scenario of the audio processor composed of various components. Based on different evaluation configurations, the audio sample data is processed respectively to obtain the intermediate files of the evaluation units under different evaluation configurations, so as to determine the audio processing effects of various components in the audio processor under different configuration scenarios. The audio quality is determined by the difference between the intermediate files of the associated evaluation units to determine whether there are large differences in the audio features processed by two adjacent components in the audio processor, and finally the functional availability of the parameters to be evaluated in the audio processor under various configurations can be determined based on the audio quality results. The present invention processes the audio sample data based on the evaluation configuration composed of the evaluation unit, and performs quality evaluation on the processed audio data, which can quickly cover and evaluate the functional availability of the components of the audio processor under all configurations, thereby improving the development efficiency and testing efficiency of the digital audio processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flowchart of a method for evaluating audio quality according to an embodiment of the present invention;

[0010] Figure 2 is a structural block diagram of a digital audio processor according to an embodiment of the present invention;

[0011] Figure 3 A flowchart of another method for evaluating audio quality according to an embodiment of the present invention;

[0012] Figure 4 A flowchart of another method for evaluating audio quality according to an embodiment of the present invention;

[0013] Figure 5A waveform diagram of analog audio data according to an embodiment of the present invention;

[0014] Figure 6 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0016] In the existing technology, during the development of digital audio processors, it is necessary to verify the audio quality processed by the internal configuration of the digital audio processor to ensure product quality. However, the digital audio processor has many internal audio interfaces, IPs and components, and the configuration is complex, so it is difficult to quickly cover and evaluate the functional availability of all configurations. In addition, the output port of the digital audio processor integrates a digital-to-analog converter, which converts the high-quality digital audio processed by the audio algorithm into analog audio, resulting in a large gap between the high-quality audio and the original audio, making it difficult to evaluate the audio quality processed by the digital-to-analog converter.

[0017] In order to solve at least the above technical problems, the present disclosure provides a method for evaluating audio quality. According to the present disclosure, based on different configuration strategies, evaluation units are combined to obtain different evaluation configurations, thereby simulating a multi-configuration scenario in which an audio processor is composed of various components. Based on different evaluation configurations, audio sample data are processed respectively to obtain intermediate files of evaluation units under different evaluation configurations, thereby determining the audio processing effects of various components in the audio processor under different configuration scenarios. The audio quality is determined by the difference between the intermediate files of the associated evaluation units to determine whether there is a large difference in the audio features processed by two adjacent components in the audio processor, and finally based on the audio quality results, the functional availability of the parameters to be evaluated in the audio processor under various configurations can be determined. In this way, according to an embodiment of the present disclosure, the audio sample data is processed based on the evaluation configuration composed of the evaluation unit, and the quality of the processed audio data is evaluated, which can quickly cover and evaluate the functional availability of the components of the audio processor under all configurations, thereby improving the development efficiency and testing efficiency of the digital audio processor.

[0018] According to an embodiment of the present disclosure, a full-link audio quality assessment process of a digital audio processor is provided.

[0019] Build an audio sample library based on audio waveforms. You can use standard audio waveforms (sine waves, white noise, etc.) generated by an audio generator or audio waveforms (sound, tone, and speech speed) generated by an artificial intelligence speech generator to build an audio sample library. The audio sample library consists of a sample list and audio files. The sample list records the audio name and the corresponding storage path. The audio chip device has a relatively small built-in storage, and generally uses an external USB flash drive to store the audio sample library.

[0020] Configure the audio quality assessment configuration. Build the minimum assessment unit based on the task parameters and audio sample files, build multiple assessment configurations based on multiple assessment units, and then build a complete quality assessment strategy. Task parameters generally refer to the combination of audio input, audio output, audio processing algorithm, and intermediate file path. The assessment unit is used to configure the audio processing algorithm. The assessment configuration generally refers to sequential testing or random testing, etc.

[0021] Perform audio quality assessment. The digital audio processor loads the task parameters of the minimum evaluation unit and performs the audio processing operations required by the evaluation unit. The audio processing algorithm is configured by the evaluation unit (for example, the combined processing algorithm of audio decoding and AI sound denoising can be configured by the evaluation unit).

[0022] Stores intermediate files of the entire audio chain. Records and stores the intermediate audio files of the digital audio processor in the three core links of audio input, audio algorithm processing, and audio output. Intermediate files are generally stored in built-in storage or external USB flash drives. The input and output of each audio processing algorithm are audio data, so the intermediate files are also audio data.

[0023] Compare the intermediate files to evaluate the audio quality. Convert the audio intermediate file to an audio histogram or perform Fourier transform, and compare the differences between the audio histograms or Fourier transform results of adjacent components to evaluate the audio quality. The smaller the difference, the higher the audio quality.

[0024] Outputs the full-link audio quality report, evaluation task list, and audio quality score.

[0025] According to an embodiment of the present disclosure, a digital DAC quality assessment process in a digital audio processor is provided.

[0026] Construct a special audio file. Use ADC simulation software or mathematical calculations to construct special audio data, and then add an audio data format header to build an audio file.

[0027] Perform audio quality assessment. Input the audio file into the digital DAC for processing, and use a high-speed oscilloscope to collect the output signal of the digital DAC. Generally, the oscilloscope is required to have a sufficiently high sampling rate. Calculate the average value of the duty cycle of the output signal corresponding to each audio file and the preset duty cycle. By comparing the duty cycle, determine whether the digital audio encoder DAC is functioning properly.

[0028] Hereinafter, the technical solution according to the present disclosure will be described with reference to specific embodiments and in conjunction with the accompanying drawings.

[0029] Figure 1 1 is a flow chart showing a method 100 for evaluating audio quality according to an embodiment of the present disclosure. Figure 1 The method 100 includes the following steps 102 to 110.

[0030] In step 102, at least two groups of parameters to be evaluated of the audio processor are obtained.

[0031] In step 104, at least two evaluation units corresponding to the at least two groups of parameters to be evaluated are respectively constructed.

[0032] In some embodiments, respectively constructing at least two evaluation units corresponding to the at least two groups of parameters to be evaluated includes: obtaining at least two audio processing units divided by the audio processor based on the system software granularity; configuring one of the audio processing units according to each group of the parameters to be evaluated to obtain at least two evaluation units. In this way, multiple audio processing units divided by the audio processor based on the system software granularity are obtained, and the audio processing unit is configured based on the parameters to be evaluated so that the audio processing unit can execute the currently configured specific parameters, thereby obtaining the corresponding intermediate file.

[0033] In step 106, the at least two evaluation units are respectively combined according to a plurality of configuration strategies to obtain a plurality of evaluation configurations.

[0034] In some embodiments, combining the at least two evaluation units according to a plurality of configuration strategies to obtain a plurality of evaluation configurations includes: combining the at least two evaluation units in a fixed order or a random order according to a plurality of configuration strategies to obtain a plurality of evaluation configurations corresponding to the plurality of configuration strategies. In some embodiments, the evaluation configuration includes a single configuration, a linear configuration, a nonlinear configuration, a composite configuration, and a random configuration. In this way, the evaluation units are combined according to the configuration strategy of the audio processor to simulate different configuration scenarios of the audio processor, and the evaluation units are combined in a fixed order or a random order, so that different evaluation units can be sequentially tested or randomly tested under different configurations in the future, thereby improving the coverage of audio quality evaluation.

[0035] In step 108, audio processing operations are respectively performed on the audio sample data based on the plurality of evaluation configurations to generate intermediate files of the at least two evaluation units under the various evaluation configurations.

[0036] In some embodiments, performing audio processing operations on audio sample data based on a plurality of evaluation configurations to generate intermediate files of the at least two evaluation units under various evaluation configurations includes: in each evaluation configuration, sequentially loading the parameters to be evaluated of the at least two evaluation units based on the combination order of the evaluation configurations, the parameters to be evaluated include parameters associated with the audio input path, the audio output path, and the audio processing algorithm; in each evaluation unit, inputting the audio sample data according to the audio input path and performing audio processing through the audio processing algorithm, and outputting the processed audio sample data according to the audio output path to generate the intermediate files of the evaluation unit under the evaluation configuration. In this way, different audio processing algorithms are respectively configured in different evaluation units, and when audio sample data is subjected to audio processing based on different configurations, the audio effects of each audio processing algorithm in different configuration scenarios can be accurately obtained, and accurate evaluation of each audio processing algorithm is achieved, while facilitating subsequent debugging of problematic audio processing algorithms, effectively improving development efficiency.

[0037] In some embodiments, generating the intermediate file of the evaluation unit under the evaluation configuration includes: recording and storing the audio data related to the audio sample data during the input, audio processing and output processes, and obtaining the intermediate file of the evaluation unit under the evaluation configuration. In this way, the audio data of the audio sample data in the three links of audio input, audio output and audio processing are recorded, so that the audio processing effect in each link can be confirmed by the difference between the audio data later, and it is convenient to quickly locate the audio processing link with problems, so as to improve the development efficiency of the audio processor.

[0038] In some embodiments, the method may further include: acquiring multiple audio sample data, and constructing an audio sample library according to the multiple audio sample data, wherein performing audio processing operations on the audio sample data based on the multiple evaluation configurations includes: performing audio processing operations on the audio sample library based on the multiple evaluation configurations. In this way, by performing audio processing operations on the audio sample library containing multiple audio sample data, multiple application scenarios of the audio processor can be effectively covered to ensure the reliability of the evaluation results.

[0039] In step 110, audio quality results of various evaluation configurations are determined according to the differences between the intermediate files of the evaluation units associated with the various evaluation configurations.

[0040] In some embodiments, according to the differences between the intermediate files of the evaluation units associated in the various evaluation configurations, determining the audio quality results of the various evaluation configurations includes: generating a corresponding audio histogram or performing a Fourier transform according to the intermediate files of the evaluation units associated in the various evaluation configurations; quantifying the audio quality of the evaluation configuration according to the differences between the audio histograms or the Fourier transform results to obtain an audio quality score, wherein the smaller the difference, the higher the audio quality score. In this way, audio features are extracted by audio histograms or Fourier transforms, and if the audio feature differences between the associated audio processing algorithms are small, it means that the two audios are closer in feature distribution, and therefore the audio quality of the audio processing algorithm is higher, that is, the audio sample data has less distortion and higher fidelity after being processed by the audio processing algorithm.

[0041] In some embodiments, after determining the audio quality results of various evaluation configurations, the method may further include: generating an audio quality report corresponding to the evaluation configuration according to the audio quality score, the audio sample data corresponding to the audio quality score, and the intermediate file. In this way, it is possible to quickly determine whether there is a problem with the current audio processor based on the audio quality score, and to quickly determine the problematic audio processing link and the possible cause of the problem based on the audio sample data and the intermediate file, so as to debug the link, thereby improving the development efficiency of the audio processor.

[0042] In some embodiments, the method may further include: generating digital audio data based on a preset target duty cycle; inputting the digital audio data into a digital-to-analog converter for processing to output analog audio data; comparing the duty cycle of the analog audio data with the target duty cycle; and evaluating the audio quality result of the digital-to-analog converter based on the comparison result. In this way, since the digital-to-analog converter converts the digital signal into an analog signal, the input audio and output audio of the digital-to-analog converter are very different, so the audio quality cannot be evaluated through an intermediate file. Therefore, based on the duty cycle characteristics of the analog signal, the specified digital audio data is first generated based on the target duty cycle, and then the analog audio data is obtained after passing through the digital-to-analog converter. If the duty cycle of the analog audio data is slightly different from the target duty cycle, it indicates that the conversion effect of the digital-to-analog converter is good, so the audio quality is high.

[0043] Figure 2 2 is a block diagram showing a structure of a digital audio processor 200 according to an embodiment of the present invention. Figure 2 The method 100 described above may be applied to a digital audio processor 200 , which includes a digital audio processing subsystem 202 , an audio input subsystem 204 , and an audio output subsystem 206 .

[0044] The digital audio processing subsystem 202 includes a DSP audio processor and a TCM module. The DSP audio processor is configured to execute DSP audio algorithms. The digital audio processing subsystem 202 also includes an AP processor configured to process user control logic of audio applications.

[0045] In some embodiments, the DSP audio algorithm includes a system audio algorithm, a user-defined audio algorithm, and an audio VAD algorithm. The DSP audio algorithm can read digital audio data from the audio input subsystem. The DSP audio algorithm can be configured by the user to start and stop the algorithm, the combination of the algorithm, and the control strategy of the algorithm.

[0046] The audio input subsystem 204 includes an ADC analog-to-digital converter, a first I2C interface, and a PDM-IN interface. The audio input subsystem 204 is configured to receive digital audio data or analog audio data from the outside. If it is digital audio data, the digital audio data is directly input into the first I2C interface and the PDM-IN interface. If it is analog audio data, the analog audio data is first input into the ADC analog-to-digital converter for conversion, and then input into the first I2C interface and the PDM-IN interface.

[0047] In some embodiments, the first I2C interface can receive audio data of the left and right channels, and the audio data is of PCM (Pulse Code Modulation) type. PCM controls the sound by adjusting the duty cycle of the signal, and it digitizes the analog signal by sampling at equal time intervals (i.e., sampling rate clock cycles). Pulse Density Modulation (PDM) is a modulation method for analog signals using digital signals, which can support up to 6-channel stereo and supports input and output dual modes.

[0048] The audio output subsystem 206 includes a PCM interface, a second I2C interface, a PDM-OUT interface, and a USB interface. The USB interface and the second I2C interface are used to carry PCM audio data. The audio interface in the audio output subsystem 206 also needs to be connected to a DAC digital-to-analog converter to output analog audio data.

[0049] In some embodiments, for a complete audio / sound effect device, the digital audio processor also needs to be configured with an analog amplifier and a loudspeaker module to produce sound.

[0050] Figure 3 3 is a flow chart showing a method 300 for evaluating audio quality according to an embodiment of the present invention. Figure 3 The method 300 evaluates the audio quality by comparing the audio intermediate files, and the method 300 includes the following steps 302 to 318.

[0051] In step 302, a standard audio waveform is generated by an audio generator or an audio waveform is generated by an artificial intelligence speech generator, and an audio sample library is constructed based on the audio waveform.

[0052] The audio sample library is a variety of test audio samples selected or produced for the currently focused application scenario. These audio samples are required to cover the current application scenario and be representative and distinctive.

[0053] In step 304, an audio processing unit of the digital audio processor is obtained. The audio processing unit is a relatively independent processing unit that can be provided in the audio system, and its division granularity depends on the software granularity of the audio system.

[0054] In step 306, each audio processing unit is configured according to each set of parameters to be evaluated, so as to construct an evaluation unit corresponding to each set of parameters to be evaluated.

[0055] In some embodiments, each set of parameters to be evaluated includes an audio input interface, a combination of an audio processing algorithm, and an audio output interface.

[0056] In step 308, a plurality of evaluation units are respectively combined according to a plurality of configuration strategies to obtain a plurality of evaluation configurations.

[0057] In some embodiments, the configuration strategy includes a single configuration strategy, a linear configuration strategy, a nonlinear configuration strategy, a composite configuration strategy, and a random configuration strategy, so the corresponding evaluation configuration includes a single configuration, a linear configuration, a nonlinear configuration, a composite configuration, and a random configuration. The evaluation configuration adopts a sequential test or a random test. Each evaluation configuration represents a complete audio quality evaluation link.

[0058] In step 310 , audio processing operations are performed on the audio sample data in the audio sample library based on each evaluation configuration.

[0059] In some embodiments, the parameters to be evaluated of each evaluation unit are loaded by a digital audio processor to perform an audio processing operation according to the parameters to be evaluated of each evaluation unit. The audio processing operation is specifically a combination of audio processing algorithms corresponding to each evaluation unit. For example, if the parameters to be evaluated of the evaluation unit are configured with an audio decoding algorithm and an AI sound denoising algorithm, then the evaluation unit includes an audio decoding operation and an AI sound denoising operation when performing the audio processing operation.

[0060] In step 312, during the execution of the audio processing operation, relevant audio data of the audio sample data in the process of audio input, audio processing and audio output are recorded and stored to obtain an intermediate file of the evaluation unit under the evaluation configuration.

[0061] In some embodiments, the intermediate files are generally stored in a built-in storage or an external USB flash drive. The intermediate files may include performance files, trace files, intermediate audio files, and information dump files related to the audio sample data.

[0062] In step 314, in each evaluation configuration, the intermediate files of the adjacent evaluation units are converted into audio histograms or Fourier transforms are performed. Adjacent evaluation units refer to two evaluation units with a direct connection between the data link and the control link.

[0063] In step 316, the audio quality of the configurations is quantitatively evaluated according to the difference between the audio histograms or Fourier transform results to obtain an audio quality score, wherein the smaller the difference, the higher the audio quality.

[0064] In step 318, an evaluation task list is first generated according to the audio sample data, the intermediate file, and the evaluation configuration type, and then an audio quality report is generated according to the evaluation task list and the corresponding audio quality scores.

[0065] In some embodiments, the audio quality report under each evaluation configuration can be characterized as a link structure. Specifically, if the first evaluation configuration is composed of evaluation unit A and evaluation unit B in sequence, then the execution link of the first evaluation configuration is evaluation unit A-evaluation unit B, and so on, the execution link of the second evaluation configuration is evaluation unit B-evaluation unit C, and the execution link of the third evaluation configuration is evaluation unit A-evaluation unit B-evaluation unit C.

[0066] In some embodiments, if there is a problem with the audio quality report of the total link, the problematic audio processing algorithm can be located through the audio quality report of the branch link. Specifically, if there is a problem with the audio quality report of the execution link AC of the third evaluation configuration, the problematic audio processing algorithm is found through the audio quality reports of the execution link AB of the first evaluation configuration and the execution link BC of the second evaluation configuration.

[0067] Figure 4 4 is a flow chart showing a method 400 for evaluating audio quality according to an embodiment of the present invention. Figure 4, the method 400 evaluates the audio quality by comparing the duty cycle. The output port of the digital audio processor integrates a DAC digital-to-analog converter, and the high-quality digital audio processed by the audio algorithm is converted into analog audio through the DAC digital-to-analog converter. The DAC digital-to-analog converter integrates audio data processing functions such as digital comparators, low-pass filters, and volume control. Since the audio data is greatly different from the input audio data after being processed by the effects of various audio algorithms of the digital audio processor, the audio data output to the DAC digital-to-analog converter is already very different from the input audio data, so it is impossible to evaluate the audio quality through the above method 300, thereby evaluating the function of the DAC digital-to-analog converter. The method 400 includes the following steps 402 to 406.

[0068] In step 402, digital audio data is generated based on a preset target duty cycle by ADC simulation software or mathematical calculation, and an audio file is constructed according to the digital audio data and a preset audio data format header.

[0069] In some embodiments, the ADC simulation software includes precision ADC online simulation tools, such as Analog Devices and ADI Precision Studio.

[0070] In some embodiments, the target duty cycles are: 20%, 40%, 60%, 80% respectively, then the ADC simulation software is configured to the corresponding operation mode and the specified reference voltage, so as to reversely calculate the input voltage corresponding to the target duty cycles of 20%, 40%, 60%, 80% respectively, so as to enable the ADC simulation software to output digital audio data with the target duty cycle by inputting a specific voltage. Specifically, the ADC simulation software is configured to the first-order Sigma-Delta mode, and its reference voltage is set to VREF=2.5V, then the input voltage can be (+ / -VREF); if the input voltage is -1.5V, the duty cycle = (-1.5+2.5) / 5 = 20%; if the input voltage is -0.5V, the duty cycle = (-0.5+2.5) / 5 = 40%; if the input voltage is 0.5V, the duty cycle = (0.5+2.5) / 5 = 60%; if the input voltage is 1.5V, the duty cycle = (1.5+2.5) / 5 = 80%. For example, when the input voltage is 1.5, the digital audio data output by the ADC simulation software is: 1 1 0 1 1 1 1 0 11 11 0 1 1 1 1 0 1 1 1 1 0 1 1 1 1 0 1 1 1 0 1 1. There are 5 0s, a total of 30, and a duty cycle of about 80%.

[0071] At step 404, the digital audio data is input to a digital-to-analog converter for processing to output analog audio data.

[0072] In some embodiments, the analog audio data output by the digital-to-analog converter is collected by a high-speed oscilloscope.

[0073] In step 406, the duty cycle of the analog audio data is compared with the target duty cycle. If the difference between the duty cycle of the analog audio data and the target duty cycle is smaller, the audio quality score of the digital-to-analog converter is higher.

[0074] In some embodiments, the four duty cycles are outputted in a relatively continuous manner with the corresponding data as follows: Figure 5 As shown, there is a transition area between each duty cycle switch, and the frequency of each duty cycle is not fixed, but fluctuates randomly. Therefore, it is difficult to perform a stable test by directly comparing the waveforms, so it cannot be introduced into mass production tests. To this end, a duty cycle value can be output in an uninterrupted loop. When capturing the duty cycle value data, ensure that the transition area has been skipped, then capture the duty cycle value data and calculate its average duty cycle to obtain the final comparison value. Compare the final comparison value with the target duty cycle to determine the audio quality score of the digital-to-analog converter.

[0075] According to another aspect of the present invention, Figure 6 is a schematic diagram showing an electronic device 500 according to an embodiment of the present invention. Figure 6 The electronic device 500 includes a memory 502, a processor 504, and a computer program stored in the memory 504 and executable on the processor. When the processor 504 executes the computer program, each step of a method for evaluating audio quality as described above is implemented.

[0076] In summary, the present invention provides a method and electronic device for evaluating audio quality. First, the audio processing unit in the digital audio processor is configured according to the parameters to be evaluated to obtain the evaluation unit. Then, the evaluation unit is combined according to different configuration strategies to obtain different evaluation configurations, thereby simulating the multi-configuration scenario in which the audio processor is composed of multiple components executing the audio processing algorithm combination. Then, the audio sample data is processed respectively based on different evaluation configurations to obtain the intermediate files of the evaluation unit under different evaluation configurations, thereby determining the audio processing effects of each component and each audio processing algorithm in the audio processor under different configuration scenarios. The audio quality is determined by the difference between the intermediate files of the associated evaluation units to determine whether there is a large difference in the audio features processed by two adjacent components in the audio processor, and finally the functional availability of the parameters to be evaluated in the audio processor under each configuration can be determined based on the audio quality results. The present invention processes the audio sample data based on the evaluation configuration composed of the evaluation unit, and performs quality evaluation on the processed audio data, which can quickly cover and evaluate the functional availability of the components of the audio processor under all configurations, thereby improving the development efficiency and testing efficiency of the digital audio processor.

[0077] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for evaluating audio quality, characterized in that: include: Obtain at least two sets of parameters to be evaluated of the audio processor; respectively constructing at least two evaluation units corresponding to the at least two groups of parameters to be evaluated; Combining the at least two evaluation units according to a plurality of configuration strategies to obtain a plurality of evaluation configurations; Based on the plurality of evaluation configurations, respectively performing audio processing operations on the audio sample data to generate intermediate files of the at least two evaluation units under the various evaluation configurations; as well as The audio quality results of the various evaluation configurations are determined according to the differences between the intermediate files of the evaluation units associated with the various evaluation configurations.

2. The method according to claim 1, characterized in that Respectively constructing at least two evaluation units corresponding to the at least two groups of parameters to be evaluated comprises: Acquire at least two audio processing units of the audio processor divided based on system software granularity; One of the audio processing units is configured according to each group of the parameters to be evaluated to obtain the at least two evaluation units.

3. The method according to claim 1, characterized in that Combining the at least two evaluation units according to various configuration strategies to obtain various evaluation configurations includes: According to a plurality of configuration strategies, the at least two evaluation units are combined in a fixed order or a random order to obtain a plurality of evaluation configurations corresponding to the plurality of configuration strategies, The evaluation configurations include single configuration, linear configuration, nonlinear configuration, composite configuration and random configuration.

4. The method according to claim 2, characterized in that: Performing audio processing operations on the audio sample data based on the multiple evaluation configurations to generate intermediate files of the at least two evaluation units under the various evaluation configurations includes: In each of the evaluation configurations, the parameters to be evaluated of the at least two evaluation units are sequentially loaded based on the combination order of the evaluation configurations, the parameters to be evaluated including parameters associated with an audio input path, an audio output path, and an audio processing algorithm; In each of the evaluation units, the audio sample data is input according to the audio input path and audio processing is performed through the audio processing algorithm, and the processed audio sample data is output according to the audio output path to generate an intermediate file of the evaluation unit under the evaluation configuration.

5. The method according to claim 4, characterized in that Generating the intermediate file of the evaluation unit under the evaluation configuration includes: Record and store relevant audio data of the audio sample data during the input, audio processing and output processes to obtain an intermediate file of the evaluation unit under the evaluation configuration.

6. The method according to claim 4, characterized in that Determining the audio quality results of various evaluation configurations according to the differences between the intermediate files of the evaluation units associated with the various evaluation configurations includes: Generate a corresponding audio histogram or perform Fourier transform according to the intermediate files of the evaluation units associated with various evaluation configurations; The audio quality of the evaluation configuration is quantified according to the difference between the audio histograms or the Fourier transform results to obtain an audio quality score, wherein the smaller the difference is, the higher the audio quality score is.

7. The method according to claim 6, characterized in that After determining the audio quality results of the various evaluation configurations, the method further includes: An audio quality report corresponding to the evaluation configuration is generated according to the audio quality score, the audio sample data corresponding to the audio quality score, and the intermediate file.

8. The method according to claim 1, characterized in that Also includes: generating digital audio data based on a preset target duty cycle; Inputting the digital audio data into a digital-to-analog converter for processing to output analog audio data; comparing the duty cycle of the analog audio data with the target duty cycle; The audio quality result of the digital-to-analog converter is evaluated according to the comparison result.

9. The method according to claim 1, characterized in that: Also includes: Acquire multiple audio sample data, and construct an audio sample library according to the multiple audio sample data, The performing of audio processing operations on the audio sample data based on the plurality of evaluation configurations includes: performing audio processing operations on the audio sample library based on the plurality of evaluation configurations.

10. An electronic device, characterized in that: include: a memory configured to store an executable program; as well as The processor is configured to execute the program so that the electronic device implements the method for evaluating audio quality according to any one of claims 1 to 9.

Images