Audio information evaluation methods, devices, electronic equipment and storage media

Abstract

This disclosure relates to an audio information evaluation method, apparatus, electronic device, and storage medium. The method includes: evaluating the singing level of an audio file to be processed to obtain a singing level evaluation result corresponding to the audio file; evaluating the recording quality of the audio file to be processed to obtain a recording quality evaluation result corresponding to the audio file; and determining an evaluation result for the audio file to be processed based on the singing level evaluation result and the recording quality evaluation result. This disclosure fully considers the inherent influence of the recording quality of the recording device, which can improve the accuracy of the evaluation results.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of audio processing, and particularly relates to an audio information evaluation method and device, electronic equipment and a storage medium. BACKGROUND

[0002] With the development of terminal technology, karaoke scoring has become a necessary function of karaoke products. Karaoke scoring mainly evaluates the karaoke audio of a user to obtain a corresponding score.

[0003] In the related art, a multi-dimensional karaoke scoring method can be used to comprehensively describe the singing level of a user. This method generally gives a comprehensive score in the dimensions of pitch, rhythm, breath, emotion, vocal range, skill, and voice.

[0004] It can be seen that the related art only evaluates the singing level of a user without considering the influence of inherent factors, resulting in insufficient accuracy of the evaluation result. SUMMARY

[0005] The present disclosure provides an audio information evaluation method and device, electronic equipment and a storage medium to at least solve the problem of insufficient accuracy of the evaluation result in the related art. The technical solutions of the present disclosure are as follows:

[0006] According to a first aspect of an embodiment of the present disclosure, an audio information evaluation method is provided, comprising:

[0007] evaluating the singing level of the to-be-processed audio to obtain a singing level evaluation result corresponding to the to-be-processed audio;

[0008] evaluating the recording quality of the to-be-processed audio to obtain a recording quality evaluation result corresponding to the to-be-processed audio;

[0009] determining an evaluation result of the to-be-processed audio according to the singing level evaluation result and the recording quality evaluation result.

[0010] Optionally, the evaluating the recording quality of the to-be-processed audio to obtain a recording quality evaluation result corresponding to the to-be-processed audio comprises:

[0011] evaluating the recording quality of the to-be-processed audio according to at least one recording quality factor to obtain an evaluation result corresponding to each recording quality factor;

[0012] performing weighted summation on the evaluation results corresponding to the at least one recording quality factor according to the weights corresponding to each recording quality factor to obtain the recording quality evaluation result.

[0013] Optionally, the recording quality factor includes a signal-to-noise ratio.

[0014] The recording quality evaluation of the to-be-processed audio according to at least one recording quality factor respectively, to obtain an evaluation result corresponding to each recording quality factor, comprises:

[0015] The to-be-processed audio is divided into a plurality of first time segments according to a first time length;

[0016] Determine the first time segment with human voice in the plurality of first time segments to obtain at least one first human voice segment;

[0017] For each first human voice segment, determine the signal-to-noise ratio of the human voice audio signal and the noise audio signal in the first human voice segment;

[0018] According to the signal-to-noise ratio of at least one first human voice segment, determine the signal-to-noise ratio evaluation result of the to-be-processed audio.

[0019] Optionally, the signal-to-noise ratio of the human voice audio signal and the noise audio signal in the first human voice segment for each first human voice segment comprises:

[0020] For each first human voice segment, blind source separation is performed on the first human voice segment to obtain the human voice audio signal in the first human voice segment;

[0021] According to the human voice audio signal, determine the human voice energy and the noise energy in the first human voice segment;

[0022] Determine the ratio of the human voice energy and the noise energy to obtain the signal-to-noise ratio of the human voice audio signal and the noise audio signal in the first human voice segment.

[0023] Optionally, the signal-to-noise ratio of the human voice audio signal and the noise audio signal in the first human voice segment for each first human voice segment comprises:

[0024] Determine the first time segment other than the first human voice segment in the plurality of first time segments as a noise segment to obtain at least one noise segment;

[0025] Determine the average value of the energy of at least one noise segment as the noise energy;

[0026] For each first human voice segment, determine the energy of the first human voice segment as the human voice energy of the first human voice segment;

[0027] Determine the ratio of the human voice energy and the noise energy to obtain the signal-to-noise ratio of the human voice audio signal and the noise audio signal in the first human voice segment.

[0028] Optionally, the determining of the signal-to-noise ratio evaluation result of the audio to be processed according to the signal-to-noise ratios of the at least one first human voice segment comprises:

[0029] determining an average value of the signal-to-noise ratios of the at least one first human voice segment to obtain a signal-to-noise ratio average value;

[0030] determining the signal-to-noise ratio evaluation result of the audio to be processed according to the signal-to-noise ratio average value, a first signal-to-noise ratio threshold value and a second signal-to-noise ratio threshold value, the first signal-to-noise ratio threshold value being smaller than the second signal-to-noise ratio threshold value.

[0031] Optionally, the recording quality factors include a bandwidth.

[0032] The recording quality evaluation of the audio to be processed according to the at least one recording quality factor to obtain an evaluation result corresponding to each recording quality factor comprises:

[0033] dividing the audio to be processed into a plurality of second time segments according to a second time length;

[0034] determining second time segments with human voice in the plurality of second time segments to obtain at least one second human voice segment;

[0035] for each second human voice segment, performing fundamental frequency detection on the second human voice segment by frame, determining frames with detected fundamental frequency as voiced signals, and determining frames without detected fundamental frequency as unvoiced signals;

[0036] determining a bandwidth of the unvoiced signals in the second human voice segment and a bandwidth of the voiced signals in the second human voice segment;

[0037] determining a bandwidth evaluation result of the audio to be processed according to the bandwidths of the unvoiced signals and the bandwidths of the voiced signals corresponding to the at least one second human voice segment.

[0038] Optionally, the determining of the bandwidth of the unvoiced signals in the second human voice segment and the bandwidth of the voiced signals in the second human voice segment comprises:

[0039] performing time-frequency transformation on each frame of unvoiced signal in the second human voice segment to obtain an amplitude spectrum of each frame of unvoiced signal, and performing time-frequency transformation on each frame of voiced signal in the second human voice segment to obtain an amplitude spectrum of each frame of voiced signal;

[0040] determining the bandwidth of the unvoiced signals according to the amplitude spectrum of each frame of unvoiced signal;

[0041] determining the bandwidth of the voiced signals according to the amplitude spectrum of each frame of voiced signal.

[0042] Optionally, the determining the bandwidth of the unvoiced signal according to the amplitude spectrum of each frame of the unvoiced signal comprises:

[0043] determining an average amplitude spectrum of the unvoiced signal according to the amplitude spectrum of each frame of the unvoiced signal as an unvoiced amplitude spectrum;

[0044] determining a maximum amplitude value in the unvoiced amplitude spectrum as an unvoiced maximum amplitude value, and determining a preset proportion of the unvoiced maximum amplitude value as an unvoiced reference amplitude value;

[0045] determining a highest frequency value in the unvoiced amplitude spectrum whose amplitude value is greater than the unvoiced reference amplitude value as a first unvoiced frequency value;

[0046] determining a frequency value in the unvoiced amplitude spectrum whose amplitude decreases most rapidly with frequency as a second unvoiced frequency value;

[0047] determining the minimum frequency value between the first unvoiced frequency value and the second unvoiced frequency value as the bandwidth of the unvoiced signal;

[0048] the determining the bandwidth of the voiced signal according to the amplitude spectrum of each frame of the voiced signal comprises:

[0049] determining an average amplitude spectrum of the voiced signal according to the amplitude spectrum of each frame of the voiced signal as a voiced amplitude spectrum;

[0050] determining a maximum amplitude value in the voiced amplitude spectrum as a voiced maximum amplitude value, and determining a preset proportion of the voiced maximum amplitude value as a voiced reference amplitude value;

[0051] determining a highest frequency value in the voiced amplitude spectrum whose amplitude value is greater than the voiced reference amplitude value as a first voiced frequency value;

[0052] determining a frequency value in the voiced amplitude spectrum whose amplitude decreases most rapidly with frequency as a second voiced frequency value;

[0053] determining the minimum frequency value between the first voiced frequency value and the second voiced frequency value as the bandwidth of the voiced signal.

[0054] Optionally, the determining the bandwidth evaluation result of the audio to be processed according to the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one of the second vocal segments comprises:

[0055] for each of the second vocal segments, determining a bandwidth evaluation result of the unvoiced signal according to the bandwidth of the unvoiced signal, a first unvoiced bandwidth threshold and a second unvoiced bandwidth threshold;

[0056] For each of the second vocal fragments, a bandwidth evaluation result of the voiced signal is determined according to the bandwidth of the voiced signal, the first voiced bandwidth threshold and the second voiced bandwidth threshold;

[0057] For each of the second vocal fragments, an average of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the voiced signal is determined as a bandwidth evaluation result of the second vocal fragment;

[0058] An average of the bandwidth evaluation results of the at least one second vocal fragment is determined as a bandwidth evaluation result of the to-be-processed audio.

[0059] Optionally, the recording quality factors include recording defects;

[0060] The recording quality evaluation of the to-be-processed audio according to the at least one recording quality factor respectively, to obtain an evaluation result corresponding to each of the recording quality factors, includes:

[0061] The to-be-processed audio is divided into a plurality of third time fragments according to a third time length;

[0062] A maximum amplitude absolute value of the to-be-processed audio is determined, and a first proportion of the maximum amplitude absolute value is determined as a first amplitude, and a second proportion of the maximum amplitude absolute value is determined as a second amplitude; the first proportion is less than the second proportion;

[0063] For each of the third time fragments, a number of data points with an amplitude absolute value between the first amplitude and the second amplitude in the third time fragment is determined as a first data point number, and a number of data points with an amplitude absolute value between the second amplitude and the maximum amplitude absolute value in the third time fragment is determined as a second data point number;

[0064] The third time fragment with the second data point number greater than the first data point number is determined as a recording defect fragment;

[0065] According to the number of the recording defect fragments and the total number of the third time fragments, a recording defect evaluation result of the to-be-processed audio is determined.

[0066] Optionally, before determining, for each of the third time fragments, the number of data points between the first amplitude and the second amplitude in the third time fragment, the method further includes:

[0067] For each of the third time fragments, a histogram of the third time fragment is determined according to the amplitude absolute value;

[0068] For each of the third time segments, determining a number of data points of the third time segment between the first amplitude value and the second amplitude value as a first data point number, comprises:

[0069] Determining a sum of the histogram between the first amplitude value and the second amplitude value as the first data point number.

[0070] Determining a number of data points of the third time segment between the second amplitude value and the maximum amplitude absolute value as a second data point number, comprises:

[0071] Determining a sum of the histogram between the second amplitude value and the maximum amplitude absolute value as the second data point number.

[0072] Optionally, determining a recording flaw evaluation result of the to-be-processed audio according to the number of recording flaw segments and a total number of the third time segments, comprises:

[0073] Determining a ratio of the number of recording flaw segments to the total number as the recording flaw evaluation result of the to-be-processed audio.

[0074] Optionally, the recording quality evaluation of the to-be-processed audio comprises:

[0075] Obtaining a device identifier and a recording mode corresponding to a recording device of the to-be-processed audio;

[0076] Obtaining a hardware score corresponding to the device identifier and the recording mode, and determining the hardware score as the recording quality evaluation result corresponding to the to-be-processed audio.

[0077] Optionally, the singing level evaluation of the to-be-processed audio comprises:

[0078] Obtaining a fundamental frequency sequence of the to-be-processed audio, and obtaining a pitch template corresponding to the to-be-processed audio;

[0079] Determining a pitch evaluation result of the to-be-processed audio according to the fundamental frequency sequence and the pitch template, and taking the pitch evaluation result as the singing level evaluation result.

[0080] Optionally, an element in the fundamental frequency sequence comprises a time point and a fundamental frequency value, and the pitch template comprises a start time, an end time and a template frequency value corresponding to a musical note.

[0081] Determining a pitch evaluation result of the to-be-processed audio according to the fundamental frequency sequence and the pitch template, comprises:

[0082] Based on the fundamental frequency value and the template frequency value, the fundamental frequency sequence is time-aligned with the tone template, and the time point corresponding to the fundamental frequency value is updated to the aligned time point;

[0083] Based on the alignment time point, determine the deviation between the fundamental frequency value and the template frequency value of each element;

[0084] The pitch evaluation result of the audio to be processed is determined based on the deviation.

[0085] Optionally, determining the deviation between the fundamental frequency value and the template frequency value of each element based on the alignment time point includes:

[0086] For each note, a target element is determined between the start time and the end time for the alignment time, and the ratio of the fundamental frequency value corresponding to the target element to the template frequency value is determined.

[0087] Based on the ratio, the deviation between the fundamental frequency value corresponding to the target element and the template frequency value is determined.

[0088] Optionally, determining the deviation between the fundamental frequency value corresponding to the target element and the template frequency value based on the ratio includes:

[0089] If the ratio is less than or equal to the first ratio threshold, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0090] If the ratio is greater than the first ratio threshold and the ratio is less than or equal to the value of the quadratic curve corresponding to the alignment time point of the target element, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value; the quadratic curve is the curve between the start time and the end time, the values ​​of the quadratic curve corresponding to the start time and the end time are respectively the second ratio threshold, the value corresponding to the center time of the quadratic curve between the start time and the end time is the first ratio threshold, and the second ratio threshold is greater than the first ratio threshold;

[0091] If the ratio is greater than the value of the quadratic curve corresponding to the alignment time point of the target element, then the ratio is determined as the deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0092] According to a second aspect of the present disclosure, an audio information evaluation apparatus is provided, comprising:

[0093] The singing level evaluation module is configured to perform a singing level evaluation on the audio to be processed and obtain the singing level evaluation result corresponding to the audio to be processed.

[0094] The recording quality evaluation module is configured to perform a recording quality evaluation on the audio to be processed and obtain the recording quality evaluation result corresponding to the audio to be processed.

[0095] The comprehensive evaluation module is configured to determine the evaluation result of the audio to be processed based on the singing level evaluation result and the recording quality evaluation result.

[0096] Optionally, the recording quality evaluation module includes:

[0097] The recording quality factor evaluation unit is configured to perform recording quality evaluation on the audio to be processed according to at least one recording quality factor, and obtain the evaluation result corresponding to each recording quality factor.

[0098] The recording quality evaluation unit is configured to perform a weighted summation of the evaluation results corresponding to at least one recording quality factor based on the weight corresponding to each of the recording quality factors, to obtain the recording quality evaluation result.

[0099] Optionally, the recording quality factors include signal-to-noise ratio;

[0100] The recording quality factor evaluation unit includes:

[0101] The first segment division subunit is configured to divide the audio to be processed into multiple first time segments according to a first time length;

[0102] The first human voice segment determination subunit is configured to perform the determination of a first time segment containing human voice among a plurality of first time segments to obtain at least one first human voice segment.

[0103] The signal-to-noise ratio determination subunit is configured to perform, for each of the first human voice segments, determining the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment;

[0104] The signal-to-noise ratio (SNR) evaluation subunit is configured to perform an SNR evaluation result of the audio to be processed based on the SNR of at least one of the first human voice segments.

[0105] Optionally, the signal-to-noise ratio determination subunit is configured to perform:

[0106] For each of the first voice segments, blind source separation is performed on the first voice segment to obtain the voice audio signal in the first voice segment;

[0107] Based on the human voice audio signal, determine the human voice energy and noise energy in the first human voice segment;

[0108] The ratio of the human voice energy to the noise energy is determined to obtain the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment.

[0109] Optionally, the signal-to-noise ratio determination subunit is configured to perform:

[0110] The first time segments other than the first human voice segment are identified as noise segments among a plurality of the first time segments, to obtain at least one noise segment;

[0111] Determine the average energy of at least one of the noise segments as the noise energy;

[0112] For each of the first voice segments, the energy of the first voice segment is determined as the voice energy of the first voice segment;

[0113] The ratio of the human voice energy to the noise energy is determined to obtain the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment.

[0114] Optionally, the signal-to-noise ratio evaluation subunit is configured to perform:

[0115] Determine the average signal-to-noise ratio of at least one of the first human voice segments to obtain the average signal-to-noise ratio;

[0116] The signal-to-noise ratio (SNR) evaluation result of the audio to be processed is determined based on the mean SNR, the first SNR threshold, and the second SNR threshold, wherein the first SNR threshold is less than the second SNR threshold.

[0117] Optionally, the recording quality factors include bandwidth;

[0118] The recording quality evaluation unit includes:

[0119] The second segment division subunit is configured to divide the audio to be processed into multiple second time segments according to a second time length;

[0120] The second voice segment determination subunit is configured to perform the determination of a second time segment containing a human voice among a plurality of second time segments, thereby obtaining at least one second voice segment;

[0121] The fundamental frequency detection subunit is configured to perform fundamental frequency detection on each frame of the second human voice segment, determine the frames in which the fundamental frequency is detected as voiced signals, and determine the frames in which the fundamental frequency cannot be detected as unvoiced signals;

[0122] The signal bandwidth determination subunit is configured to perform the determination of the bandwidth of the unvoiced signal in the second voice segment and the determination of the bandwidth of the voiced signal in the second voice segment;

[0123] The bandwidth evaluation subunit is configured to perform a bandwidth evaluation result of the audio to be processed based on the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one of the second vocal segments.

[0124] Optionally, the signal bandwidth determination subunit includes:

[0125] The time-frequency transformation submodule is configured to perform time-frequency transformation on each frame of unvoiced sound signal in the second voice segment to obtain the amplitude spectrum of each frame of unvoiced sound signal, and to perform time-frequency transformation on each frame of voiced sound signal in the second voice segment to obtain the amplitude spectrum of each frame of voiced sound signal.

[0126] The unvoiced bandwidth determination submodule is configured to determine the bandwidth of the unvoiced signal based on the amplitude spectrum of the unvoiced signal in each frame;

[0127] The voiced signal bandwidth determination submodule is configured to determine the bandwidth of the voiced signal based on the amplitude spectrum of the voiced signal in each frame.

[0128] Optionally, the unvoiced bandwidth determination submodule is configured to perform:

[0129] Based on the amplitude spectrum of the unvoiced signal in each frame, the average amplitude spectrum of the unvoiced signal is determined as the unvoiced amplitude spectrum;

[0130] The maximum amplitude value in the unvoiced tone amplitude spectrum is determined as the maximum amplitude value of the unvoiced tone, and a preset proportion of the maximum amplitude value of the unvoiced tone is determined as the reference amplitude value of the unvoiced tone.

[0131] The highest frequency value in the unvoiced tone amplitude spectrum whose amplitude value is greater than the unvoiced tone reference amplitude value is determined as the first unvoiced tone frequency value;

[0132] The frequency value at which the amplitude of the unvoiced tone decreases the fastest with frequency in the unvoiced tone amplitude spectrum is determined as the second unvoiced tone frequency value;

[0133] The frequency value that is the minimum of the first and second unvoiced frequency values ​​is determined as the bandwidth of the unvoiced signal;

[0134] The voiced bandwidth determination submodule is configured to execute:

[0135] Based on the amplitude spectrum of the voiced signal in each frame, the average amplitude spectrum of the voiced signal is determined as the voiced amplitude spectrum;

[0136] The maximum amplitude value in the voiced amplitude spectrum is determined as the maximum amplitude value of the voiced sound, and a preset proportion of the maximum amplitude value of the voiced sound is determined as the reference amplitude value of the voiced sound.

[0137] The highest frequency value in the voiced amplitude spectrum whose amplitude value is greater than the voiced reference amplitude value is determined as the first voiced frequency value;

[0138] The frequency value at which the amplitude of the voiced sound amplitude spectrum decreases the fastest with frequency is determined as the second voiced sound frequency value;

[0139] The frequency value that is the minimum of the first voiced frequency value and the second voiced frequency value is determined as the bandwidth of the voiced signal.

[0140] Optionally, the bandwidth evaluation subunit is configured to perform:

[0141] For each of the second human voice segments, the bandwidth evaluation result of the unvoiced signal is determined based on the bandwidth of the unvoiced signal, the first unvoiced bandwidth threshold, and the second unvoiced bandwidth threshold.

[0142] For each of the second human voice segments, the bandwidth evaluation result of the voiced signal is determined based on the bandwidth of the voiced signal, the first voiced bandwidth threshold, and the second voiced bandwidth threshold.

[0143] For each second voice segment, the average of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the voiced signal is determined as the bandwidth evaluation result of the second voice segment;

[0144] The average value of the bandwidth evaluation results of at least one of the second human voice segments is determined as the bandwidth evaluation result of the audio to be processed.

[0145] Optionally, the recording quality factors include recording defects;

[0146] The recording quality evaluation unit includes:

[0147] The third segment division subunit is configured to divide the audio to be processed into multiple third time segments according to a third time length;

[0148] An amplitude determination subunit is configured to perform the following operations: determine the maximum absolute amplitude value of the audio to be processed, determine a first proportion of the maximum absolute amplitude value as a first amplitude, and determine a second proportion of the maximum absolute amplitude value as a second amplitude; the first proportion is less than the second proportion.

[0149] The data point determination subunit is configured to perform, for each of the third time segments, determine the number of data points in the third time segment whose absolute amplitude value is between the first amplitude value and the second amplitude value, as the first data point count, and determine the number of data points in the third time segment whose absolute amplitude value is between the second amplitude value and the absolute value of the maximum amplitude value, as the second data point count;

[0150] The defective segment determination subunit is configured to determine a third time segment in which the number of the second data points is greater than the number of the first data points as a defective segment to be recorded;

[0151] The recording defect evaluation subunit is configured to determine the recording defect evaluation result of the audio to be processed based on the number of recording defect segments and the total number of the third time segments.

[0152] Optionally, the recording quality evaluation unit further includes:

[0153] The histogram-determining subunit is configured to perform a histogram determination for each of the third time segments based on the absolute value of the amplitude.

[0154] The data point determination subunit is configured to execute:

[0155] For each of the third time segments, the sum of the histogram between the first amplitude and the second amplitude is determined as the first data point count; the sum of the histogram between the second amplitude and the absolute value of the maximum amplitude is determined as the second data point count.

[0156] Optionally, the recording defect evaluation subunit is configured to perform:

[0157] The ratio of the number of defective recording segments to the total number is determined as the recording defect evaluation result of the audio to be processed.

[0158] Optionally, the recording quality evaluation module includes:

[0159] The device information acquisition unit is configured to acquire the device identifier and recording mode corresponding to the recording device of the audio to be processed;

[0160] The recording quality evaluation result determination unit is configured to acquire the hardware score corresponding to the device identifier and recording mode, and determine the hardware score as the recording quality evaluation result corresponding to the audio to be processed.

[0161] Optionally, the singing level evaluation module includes:

[0162] The fundamental frequency information acquisition unit is configured to acquire the fundamental frequency sequence of the audio to be processed and acquire the tone template corresponding to the audio to be processed;

[0163] The singing level evaluation unit is configured to perform a pitch evaluation result of the audio to be processed based on the fundamental frequency sequence and the pitch template, and use the pitch evaluation result as the singing level evaluation result.

[0164] Optionally, the elements in the fundamental frequency sequence include time points and fundamental frequency values, and the pitch template includes a start time, an end time, and a template frequency value corresponding to the note;

[0165] The singing level evaluation unit includes:

[0166] The time alignment subunit is configured to perform time alignment of the fundamental frequency sequence with the tone template based on the fundamental frequency value and the template frequency value, and update the time point corresponding to the fundamental frequency value to the alignment time point;

[0167] The frequency deviation determination subunit is configured to perform the following: determine the deviation between the fundamental frequency value and the template frequency value of each element based on the alignment time point.

[0168] The pitch evaluation subunit is configured to perform a pitch evaluation result of the audio to be processed based on the deviation.

[0169] Optionally, the frequency deviation determination subunit includes:

[0170] The frequency ratio determination submodule is configured to perform, for each of the notes, determine a target element between the start time and the end time for the alignment time, and determine the ratio of the fundamental frequency value corresponding to the target element to the template frequency value;

[0171] The frequency deviation determination submodule is configured to perform the operation of determining the deviation between the fundamental frequency value corresponding to the target element and the template frequency value based on the ratio.

[0172] Optionally, the frequency deviation determination submodule is configured to perform:

[0173] If the ratio is less than or equal to the first ratio threshold, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0174] If the ratio is greater than the first ratio threshold and the ratio is less than or equal to the value of the quadratic curve corresponding to the alignment time point of the target element, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value; the quadratic curve is the curve between the start time and the end time, the values ​​of the quadratic curve corresponding to the start time and the end time are respectively the second ratio threshold, the value corresponding to the center time of the quadratic curve between the start time and the end time is the first ratio threshold, and the second ratio threshold is greater than the first ratio threshold;

[0175] If the ratio is greater than the value of the quadratic curve corresponding to the alignment time point of the target element, then the ratio is determined as the deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0176] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0177] processor;

[0178] Memory used to store the processor's executable instructions;

[0179] The processor is configured to execute the instructions to implement the audio information evaluation method as described in the first aspect.

[0180] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein when instructions in the computer storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the audio information evaluation method as described in the first aspect.

[0181] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program or computer instructions, characterized in that, when the computer program or computer instructions are executed by a processor, they implement the audio information evaluation method described in the first aspect.

[0182] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0183] This embodiment of the disclosure determines the final evaluation result of the audio to be processed by combining the singing level evaluation result and the recording quality evaluation result. It fully considers the influence of the inherent factor of the recording quality of the recording device, which can improve the accuracy of the evaluation result.

[0184] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0185] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0186] Figure 1 This is a flowchart illustrating an audio information evaluation method according to an exemplary embodiment;

[0187] Figure 2 This is a schematic diagram of a quadratic curve in an embodiment of this disclosure;

[0188] Figure 3 This is a flowchart illustrating an audio information evaluation method according to an exemplary embodiment;

[0189] Figure 4 This is a flowchart illustrating the evaluation of recording quality of the audio to be processed based on the signal-to-noise ratio in this embodiment of the disclosure;

[0190] Figure 5 This is a flowchart illustrating the recording quality evaluation of the audio to be processed based on bandwidth in this embodiment of the disclosure;

[0191] Figure 6 This is a flowchart illustrating the recording quality evaluation process for audio files with recording defects in this embodiment of the present disclosure.

[0192] Figure 7 This is a block diagram illustrating an audio information evaluation device according to an exemplary embodiment;

[0193] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0194] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0195] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure 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 disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0196] Figure 1This is a flowchart illustrating an audio information evaluation method according to an exemplary embodiment, such as... Figure 1 As shown, this audio information evaluation method can be used in electronic devices such as mobile phones, tablets, computers, and servers, and includes the following steps.

[0197] In step S11, the singing level of the audio to be processed is evaluated to obtain the singing level evaluation result corresponding to the audio to be processed.

[0198] The audio to be processed is the user's karaoke audio, which can be the entire song sung by the user, or a sentence or word from the song. This embodiment of the disclosure involves scoring the audio to be processed; the audio information to be determined is the evaluation result of the audio to be processed, that is, the evaluation score of the audio to be processed.

[0199] The evaluation of the singing level of the processed audio mainly involves evaluating the singing level dimensions of the processed audio, such as pitch, rhythm, breath control, emotion, vocal range, technique, and voice, and giving a comprehensive score for at least one singing level dimension to obtain the singing level evaluation result corresponding to the processed audio.

[0200] In step S12, the recording quality of the audio to be processed is evaluated to obtain the recording quality evaluation result corresponding to the audio to be processed.

[0201] The singing level and recording quality (quality of the recording equipment) of the audio to be processed will both affect the evaluation results. For example, a muffled tone may be caused by insufficient breath support during singing or by the recording equipment's inability to record high-frequency signals; a low pitch score may be due to the user singing off-key or by current noise interference from the recording equipment. Therefore, this embodiment combines the evaluation results of singing level and recording quality of the audio to be processed to give a final evaluation result.

[0202] When evaluating the recording quality of the audio to be processed, the recording quality can be evaluated according to pre-set recording quality factors related to recording quality to obtain the recording quality evaluation result corresponding to the audio to be processed; alternatively, the recording quality evaluation result corresponding to the audio to be processed can be determined based on the recording device of the audio to be processed.

[0203] In step S13, the evaluation result of the audio to be processed is determined based on the singing level evaluation result and the recording quality evaluation result.

[0204] The evaluation result of the audio to be processed is determined by combining the singing level evaluation result and the recording quality evaluation result. This can be achieved by weighted summation of the singing level evaluation result and the recording quality evaluation result based on pre-set weights; alternatively, other methods can be used to combine the singing level evaluation result and the recording quality evaluation result, such as multinomial fitting.

[0205] In an exemplary embodiment, determining the evaluation result of the audio to be processed based on the singing level evaluation result and the recording quality evaluation result includes: performing polynomial fitting on the singing level evaluation result and the recording quality evaluation result to obtain the evaluation result of the audio to be processed.

[0206] The singing level evaluation results and recording quality evaluation results are fused into the final evaluation result using a polynomial fitting method. The evaluation result of the audio to be processed is determined according to the following formula:

[0207] S=a0A 0 +a1A 1 +…+a n A n +b0B 0 +b1B 1 +…+b m B m

[0208] Where S represents the evaluation result of the audio to be processed, A represents the singing level evaluation result, B represents the recording quality evaluation result, and a0, a1, ..., a n And b0, b1, ..., b m These are the polynomial coefficients. n and m can be preset.

[0209] The polynomial coefficients mentioned above can be obtained by comprehensively scoring singing audios with different singing levels from multiple professional evaluators and fitting the polynomial. For example, the process of determining the polynomial coefficients may include: acquiring multiple singing audios with different singing levels; acquiring comprehensive scores from multiple vocal professional evaluators for each singing audio, and using the average of the comprehensive scores for the same singing audio as the evaluation result; determining the singing level evaluation result and recording quality evaluation result for each singing audio; establishing a polynomial to fit the recording quality evaluation result and singing level evaluation result of the singing audio into the evaluation result of the singing audio, and minimizing the mean square error through a fitting algorithm to obtain the polynomial coefficients.

[0210] By combining the comprehensive scores given by professional vocal evaluators to determine the polynomial coefficients, these coefficients can be used to perform polynomial fitting on the singing level evaluation results and recording quality evaluation results for each audio data to be processed, thus obtaining the final evaluation result. The evaluation result obtained in this way is consistent with the results given by professional evaluators, improving the accuracy of the evaluation result.

[0211] The audio information evaluation method provided in this exemplary embodiment determines the final evaluation result of the audio to be processed by combining the singing level evaluation result and the recording quality evaluation result. It fully considers the influence of the inherent factor of the recording quality of the recording device and can improve the accuracy of the evaluation result.

[0212] In an exemplary embodiment, the step of evaluating the recording quality of the audio to be processed to obtain the recording quality evaluation result corresponding to the audio to be processed includes: obtaining the device identifier and recording mode corresponding to the recording device of the audio to be processed; obtaining the hardware score corresponding to the device identifier and recording mode, and determining the hardware score as the recording quality evaluation result corresponding to the audio to be processed.

[0213] The recording quality rating can be achieved by testing the recording quality of existing audio recording devices (hardware devices) in specific recording modes, obtaining the device identifier and the hardware score corresponding to the recording mode, and storing these scores in a database. When evaluating the recording quality of the audio to be processed, the device identifier and recording mode of the recording device used by the user are collected, and the hardware score corresponding to that device identifier and recording mode is retrieved from the database. This hardware score is then used as the recording quality evaluation result for the audio to be processed. The recording mode may include standard mode, conference mode, etc.

[0214] By pre-evaluating the recording quality of various recording devices and modes and storing the results in a database, the corresponding evaluation results can be directly retrieved from the database when evaluating the recording quality of the audio to be processed, thus improving the processing efficiency of recording quality evaluation.

[0215] In an exemplary embodiment, the step of evaluating the singing level of the audio to be processed to obtain the singing level evaluation result corresponding to the audio to be processed includes: obtaining the fundamental frequency sequence of the audio to be processed and obtaining the pitch template corresponding to the audio to be processed; determining the pitch evaluation result of the audio to be processed based on the fundamental frequency sequence and the pitch template, and using the pitch evaluation result as the singing level evaluation result.

[0216] The audio to be processed can be user-generated raw audio, which is pure human voice without music, meaning it is recorded without any spatial or post-processing effects. The pitch template is a preset pitch template corresponding to the song corresponding to the audio to be processed.

[0217] The fundamental frequency of the audio to be processed can be extracted using either time-domain or frequency-domain methods to obtain the fundamental frequency sequence. The song identifier corresponding to the audio is then obtained, along with the corresponding pitch template. The fundamental frequency sequence is matched with the pitch template, and the matching result is converted into a quantized score according to a certain strategy to obtain the pitch evaluation result of the audio. This pitch evaluation result is then used as the performance level evaluation result.

[0218] It should be noted that, in addition to using pitch accuracy as the evaluation result for singing level, pitch accuracy can also be combined with at least one of the following dimensions: rhythm, breath control, emotion, vocal range, technique, and voice, to obtain a more accurate evaluation result for singing level.

[0219] The pitch evaluation result is determined based on the fundamental frequency sequence and pitch template of the audio to be processed, and the pitch evaluation result is used as the singing level evaluation result. The evaluation method is relatively simple and the processing speed is fast, which can improve the efficiency of singing level evaluation.

[0220] Based on the above technical solution, the elements in the fundamental frequency sequence include time points and fundamental frequency values, and the pitch template includes the start time, end time, and template frequency value corresponding to the note;

[0221] Determining the pitch evaluation result of the audio to be processed based on the fundamental frequency sequence and the pitch template includes: aligning the fundamental frequency sequence and the pitch template in time according to the fundamental frequency value and the template frequency value, and updating the time point corresponding to the fundamental frequency value to the aligned time point; determining the deviation between the fundamental frequency value and the template frequency value of each element based on the aligned time point; and determining the pitch evaluation result of the audio to be processed based on the deviation.

[0222] The fundamental frequency sequence is composed of multiple elements (each element being a data point in the fundamental frequency sequence), each element including a time point and a fundamental frequency value. The pitch template can be in MIDI (Musical Instrument Digital Interface) format and contains a sequence of notes. Each note contains information including, but not limited to, a start time, an end time, and a template frequency value. In the pitch template, a note has a fixed template frequency value, which lasts for the duration between the start and end times. The fundamental frequency value of the same note in the fundamental frequency sequence fluctuates around the template frequency value.

[0223] Based on the fundamental frequency value in the fundamental frequency sequence and the template frequency value in the pitch template, the fundamental frequency sequence and the pitch template are time-aligned. A dynamic time warping algorithm can be used to stretch the time length of multiple elements in the fundamental frequency sequence corresponding to the same note, obtaining the time mapping relationship between the fundamental frequency sequence and the pitch template. Based on this time mapping relationship, the time point of each element in the fundamental frequency sequence is updated to the aligned time point. Based on the aligned time point, the deviation between the fundamental frequency value of each element in the fundamental frequency sequence and the template frequency value at the same aligned time point is determined. The deviation corresponding to each element is statistically analyzed, and the statistical value of the deviation is determined as the pitch evaluation result of the audio to be processed. The statistical value can be the mean or standard deviation, etc.

[0224] By aligning the fundamental frequency sequence with the pitch template in time, the deviation between the fundamental frequency value and the template frequency value is determined. This deviation is more accurate, and the pitch evaluation result of the audio to be processed can be determined based on the deviation, which can improve the accuracy of the pitch evaluation result.

[0225] Based on the above technical solution, determining the deviation between the fundamental frequency value and the template frequency value of each element according to the alignment time point includes:

[0226] For each note, a target element is determined between the start time and the end time for the alignment time, and the ratio of the fundamental frequency value corresponding to the target element to the template frequency value is determined.

[0227] Based on the ratio, the deviation between the fundamental frequency value corresponding to the target element and the template frequency value is determined.

[0228] For each note, the fundamental frequency value in the fundamental frequency sequence is compared with the template frequency value in the pitch template. Based on the start and end times of each note in the pitch template, elements in the audio sequence whose alignment time points fall between the start and end times are identified as target elements. For each target element, the ratio of its corresponding fundamental frequency value to the template frequency value is determined. When determining the ratio, the larger frequency value between the fundamental and template frequencies is used as the numerator, and the smaller frequency value is used as the denominator. After obtaining the ratio of the fundamental frequency value to the template frequency value, the deviation between the fundamental frequency value and the template frequency value corresponding to each target element can be determined based on the relationship between this ratio and a pre-set threshold.

[0229] Since the fundamental frequency value of each note fluctuates based on the template frequency value, by determining the ratio of the fundamental frequency value of each target element to the template frequency value, and then determining the deviation between the fundamental frequency value of the target element and the template frequency value based on this ratio, the accuracy of deviation determination can be further improved, thereby improving the accuracy of pitch evaluation results.

[0230] Based on the above technical solution, determining the deviation between the fundamental frequency value corresponding to the target element and the template frequency value according to the ratio includes:

[0231] If the ratio is less than or equal to the first ratio threshold, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0232] If the ratio is greater than the first ratio threshold and the ratio is less than or equal to the value of the quadratic curve corresponding to the alignment time point of the target element, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value; the quadratic curve is the curve between the start time and the end time, the values ​​of the quadratic curve corresponding to the start time and the end time are respectively the second ratio threshold, the value corresponding to the center time of the quadratic curve between the start time and the end time is the first ratio threshold, and the second ratio threshold is greater than the first ratio threshold;

[0233] If the ratio is greater than the value of the quadratic curve corresponding to the alignment time point of the target element, then the ratio is determined as the deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0234] The first ratio threshold is a value greater than 1 and less than 1.5, the second ratio threshold is a value greater than 1 and less than 1.5, and the second ratio threshold is greater than the first ratio threshold. For example, the first ratio threshold can be 1.059 and the second ratio threshold can be 1.25.Figure 2 This is a schematic diagram of a quadratic curve in an embodiment of this disclosure, such as... Figure 2 As shown, the quadratic curve is determined based on the start time, end time, first ratio threshold, and second ratio threshold. When determining the quadratic curve, the center time between the start time and the end time is determined, the second ratio threshold c2 is determined as the value of the quadratic curve at the start time and the end time, and the first ratio threshold c1 is determined as the value of the quadratic curve at the center time.

[0235] First, compare the ratio with the first ratio threshold. If the ratio is less than the first ratio threshold, it indicates that the fundamental frequency value fluctuates around the template frequency value, and the difference between the two is not significant. In this case, it can be determined that there is no deviation between the fundamental frequency value of the target element and the template frequency value. For target elements with a ratio greater than the first ratio threshold, further compare it with the value of the quadratic curve. If the ratio corresponding to the target element is less than or equal to the value of the quadratic curve corresponding to the alignment time point of the target element, it is determined that there is no deviation between the fundamental frequency value of the target element and the template frequency value. If the ratio corresponding to the target element is greater than the value of the quadratic curve corresponding to the alignment time point of the target element, it is determined that there is a certain deviation between the fundamental frequency value of the target element and the template frequency value, and this ratio is determined as the deviation between the fundamental frequency value of the target element and the template frequency value.

[0236] By setting a quadratic curve based on the start and end times of the notes, and comparing the ratio corresponding to the target element with the value of the quadratic curve at the alignment time point, the deviation between the fundamental frequency value and the template frequency value is determined based on the comparison result. This fully considers the fluctuation range of the fundamental frequency value, which can improve the accuracy of the determined deviation and further improve the accuracy of the pitch evaluation result.

[0237] Figure 3 This is a flowchart illustrating an audio information evaluation method according to an exemplary embodiment, such as... Figure 3 As shown, the audio information evaluation method includes the following steps.

[0238] In step S31, the singing level of the audio to be processed is evaluated to obtain the singing level evaluation result corresponding to the audio to be processed.

[0239] In step S32, the audio to be processed is evaluated for recording quality according to at least one recording quality factor, and the evaluation result corresponding to each recording quality factor is obtained.

[0240] Among them, recording quality factors are those that affect recording quality, such as signal-to-noise ratio, audio signal bandwidth, recording defects, etc.

[0241] For each recording quality factor, the audio to be processed is processed accordingly, and the recording quality of that recording quality factor is evaluated to obtain the evaluation result for each recording quality factor.

[0242] In step S33, the evaluation results corresponding to at least one recording quality factor are weighted and summed according to the weight corresponding to each recording quality factor to obtain the recording quality evaluation result.

[0243] The weights of each recording quality factor can be pre-set according to requirements. Then, after evaluating the recording quality of the audio to be processed based on at least one recording quality factor and obtaining the evaluation results for each factor, the evaluation results for each factor can be weighted and summed to obtain the final recording quality evaluation result. For example, three recording quality factors (signal-to-noise ratio, audio signal bandwidth, and recording defects) can be used to evaluate the recording quality of the audio to be processed. The weighting ratio of these three factors can be 1:1:1, meaning each factor has a weight of 1 / 3. After evaluating the recording quality of the audio to be processed based on the signal-to-noise ratio, audio signal bandwidth, and recording defects respectively, the evaluation results for these three factors are weighted and summed to obtain the final recording quality evaluation result for the audio to be processed.

[0244] In step S34, the evaluation result of the audio to be processed is determined based on the singing level evaluation result and the recording quality evaluation result.

[0245] The audio information evaluation method provided in this exemplary embodiment evaluates the recording quality of the audio to be processed according to at least one recording quality factor, obtains the evaluation result corresponding to each recording quality factor, and then performs a weighted summation of the evaluation results corresponding to at least one recording quality factor to obtain the recording quality evaluation result. This method combines the evaluation results of at least one recording quality factor to obtain the final recording quality evaluation result, which can improve the accuracy of the recording quality evaluation result.

[0246] In one exemplary embodiment, the recording quality factor includes the signal-to-noise ratio; Figure 4 This is a flowchart illustrating the evaluation of recording quality of the audio to be processed based on the signal-to-noise ratio in this embodiment of the disclosure, such as... Figure 4 As shown, the step of evaluating the recording quality of the audio to be processed according to at least one recording quality factor, and obtaining the evaluation result corresponding to each recording quality factor, includes:

[0247] In step S41, the audio to be processed is divided into multiple first time segments according to the first time length.

[0248] The first time length is a pre-set time segment.

[0249] The audio to be processed is divided according to the first time length, resulting in multiple first time segments corresponding to the audio to be processed.

[0250] In step S42, a first time segment containing human voice is determined from a plurality of the first time segments to obtain at least one first human voice segment.

[0251] By performing voice extraction algorithms or models on each first time segment, the first time segment from which voice can be extracted is determined, resulting in a first time segment containing voice. This first time segment with voice is then used as the first voice segment, thus obtaining at least one first voice segment. First time segments from which no voice can be extracted are considered noise segments.

[0252] In step S43, for each of the first human voice segments, the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment is determined.

[0253] After excluding noise segments, the signal-to-noise ratio (SNR) is calculated for the first human voice segment. For each first human voice segment, the human voice audio signal and the noise audio signal in the segment are determined, and the ratio of the human voice energy corresponding to the human voice audio signal to the noise energy corresponding to the noise audio signal is calculated to obtain the SNR of the human voice audio signal and the noise audio signal in the first human voice segment.

[0254] In an exemplary embodiment, determining the signal-to-noise ratio (SNR) of the human voice audio signal to the noise audio signal in each of the first human voice segments includes: performing blind source separation on each of the first human voice segments to obtain the human voice audio signal in the first human voice segment; determining the human voice energy and noise energy in the first human voice segment based on the human voice audio signal; and determining the ratio of the human voice energy to the noise energy to obtain the SNR of the human voice audio signal to the noise audio signal in the first human voice segment.

[0255] Blind source separation involves estimating the source signal based solely on the observed mixed signal without knowing the source signal or the signal mixing parameters.

[0256] The entire first voice segment is used as a mixed signal, and the voice audio signal within the first voice segment is used as the source signal. The first voice segment includes both voice audio signal and noise audio signal. By performing blind source separation on the first voice segment, the voice audio signal within the first voice segment is obtained. The energy corresponding to the voice audio signal in the first voice segment can then be determined, thus obtaining the voice energy in the first voice segment. The noise energy corresponding to the noise audio signal in the first voice segment is obtained by subtracting the voice energy from the total energy of the first voice segment. The ratio of voice energy to noise energy is then determined, resulting in the signal-to-noise ratio (SNR) of the voice audio signal to the noise audio signal in the first voice segment. Blind source separation can be performed, for example, using an Independent Component Analysis (ICA) algorithm.

[0257] By performing blind source separation on the first human voice segment to separate the human voice audio signal in the first human voice segment, more accurate human voice energy and noise energy can be obtained, thereby improving the accuracy of the signal-to-noise ratio.

[0258] In another exemplary embodiment, determining the signal-to-noise ratio (SNR) of the human voice audio signal to the noise audio signal in each of the first human voice segments includes: identifying first time segments other than the first human voice segments from a plurality of first time segments as noise segments to obtain at least one noise segment; determining the average energy of the at least one noise segment as the noise energy; determining the energy of the first human voice segment as the human voice energy of the first human voice segment for each of the first human voice segments; and determining the ratio of the human voice energy to the noise energy to obtain the SNR of the human voice audio signal to the noise audio signal in the first human voice segment.

[0259] Among multiple first time segments, all first time segments excluding the first human voice segment are identified as noise segments; that is, first time segments without human voice are identified as noise segments, resulting in at least one noise segment. The energy of each noise segment is calculated, and the average energy of at least one noise segment is determined as the noise energy of the noise audio signal in the first human voice segment. The energy of each first human voice segment is determined separately, and the energy of each first human voice segment is taken as the human voice energy of the human voice audio signal in that first human voice segment. For each first human voice segment, the ratio of human voice energy to noise energy is calculated to obtain the signal-to-noise ratio (SNR) of the human voice audio signal to the noise audio signal in each first human voice segment.

[0260] By using the average energy of at least one noise segment as the noise energy and the energy of the first human voice segment as the human voice energy, the computational load is low, which can improve the efficiency of determining the signal-to-noise ratio.

[0261] In step S44, the signal-to-noise ratio (SNR) evaluation result of the audio to be processed is determined based on the SNR of at least one of the first human voice segments.

[0262] The signal-to-noise ratio (SNR) of at least one first human voice segment is statistically analyzed, and the SNR evaluation result of the audio to be processed is determined based on the statistical results.

[0263] In an exemplary embodiment, determining the signal-to-noise ratio (SNR) evaluation result of the audio to be processed based on the SNR of at least one of the first human voice segments includes: determining the average SNR of at least one of the first human voice segments to obtain an average SNR value; and determining the SNR evaluation result of the audio to be processed based on the average SNR value, a first SNR threshold, and a second SNR threshold, wherein the first SNR threshold is less than the second SNR threshold.

[0264] Calculate the average signal-to-noise ratio (SNR) of at least one first human voice segment to obtain the mean SNR. Set two different SNR thresholds, with the first SNR threshold being less than the second SNR threshold. Compare the mean SNR with both the first and second SNR thresholds. If the mean SNR is less than or equal to the first SNR threshold, the SNR evaluation result is determined to be 0 (i.e., the SNR evaluation score is 0). If the mean SNR is greater than or equal to the second SNR threshold, the SNR evaluation result is determined to be a perfect score. If the mean SNR is greater than the first SNR threshold but less than the second SNR threshold, the SNR evaluation result is determined according to the following formula:

[0265]

[0266] Among them, S snr S represents the signal-to-noise ratio evaluation result. Fsnt This represents the maximum score for signal-to-noise ratio. SNR represents the mean signal-to-noise ratio. th1 Represents the first signal-to-noise ratio threshold, SNR th2 This represents the second signal-to-noise ratio threshold.

[0267] By determining the average signal-to-noise ratio (SNR) of at least one first human voice segment, and by comparing the average SNR with a first SNR threshold and a second SNR threshold, a more accurate SNR evaluation result can be obtained, thereby improving the accuracy of the SNR evaluation.

[0268] By dividing the audio to be processed into multiple first time segments, determining at least one first voice segment among the multiple first time segments, determining the signal-to-noise ratio (SNR) of each first voice segment, and then statistically analyzing the SNR of at least one first voice segment, the SNR evaluation result of the audio to be processed can be obtained, thus realizing the SNR evaluation of the audio to be processed.

[0269] In one exemplary embodiment, the recording quality factor includes bandwidth; Figure 5 This is a flowchart illustrating the recording quality evaluation of the audio to be processed based on bandwidth in this embodiment of the disclosure, such as... Figure 5 As shown, the step of evaluating the recording quality of the audio to be processed according to at least one recording quality factor, and obtaining the evaluation result corresponding to each recording quality factor, includes:

[0270] In step S51, the audio to be processed is divided into multiple second time segments according to the second time length.

[0271] The second time length is a pre-set duration of a second time segment. This second time length can be the same as or different from the first time length.

[0272] The audio to be processed is divided according to the second time length, resulting in multiple second time segments corresponding to the audio to be processed.

[0273] In step S52, a second time segment containing human voice is determined from among a plurality of the second time segments to obtain at least one second human voice segment.

[0274] By performing voice extraction on each second time segment using a voice extraction algorithm or model, the second time segment from which voice can be extracted is determined, and the second time segment containing voice is obtained. The second time segment containing voice is then used as the second voice segment, resulting in at least one second voice segment.

[0275] In step S53, for each of the second voice segments, the fundamental frequency of the second voice segment is detected frame by frame. Frames in which the fundamental frequency is detected are determined as voiced signals, and frames in which the fundamental frequency cannot be detected are determined as unvoiced signals.

[0276] For each second voice segment, fundamental frequency detection is performed on each frame of the second voice segment. Frames where the fundamental frequency cannot be detected are identified as unvoiced signals and marked with an unvoiced signal flag (e.g., the unvoiced signal flag can be 0). Frames where the fundamental frequency can be detected are identified as voiced signals and marked with a voiced signal flag (e.g., the voiced signal flag can be 1). Based on a pre-constrained fundamental frequency range (typically 80Hz to 1100Hz), the unvoiced signals identified based on fundamental frequency detection are updated. Voiced signals outside the fundamental frequency range are changed to unvoiced signals, and the voiced signal flag is changed to an unvoiced signal flag.

[0277] In step S54, the bandwidth of the unvoiced signal in the second voice segment is determined, and the bandwidth of the voiced signal in the second voice segment is also determined.

[0278] The bandwidth of the unvoiced signal and the bandwidth of the voiced signal can be determined by performing a video transformation on the second voice segment and based on the transformation result.

[0279] In an exemplary embodiment, determining the bandwidth of the unvoiced signal in the second voice segment and determining the bandwidth of the voiced signal in the second voice segment includes: performing a time-frequency transformation on each frame of the unvoiced signal in the second voice segment to obtain the amplitude spectrum of each frame of the unvoiced signal, and performing a time-frequency transformation on each frame of the voiced signal in the second voice segment to obtain the amplitude spectrum of each frame of the voiced signal; determining the bandwidth of the unvoiced signal based on the amplitude spectrum of each frame of the unvoiced signal; and determining the bandwidth of the voiced signal based on the amplitude spectrum of each frame of the voiced signal.

[0280] Time-frequency transformation can be performed on all second voice segments frame by frame. Assuming a second voice segment has N frames, there will be N frames after transformation. Each frame consists of M complex frequency values ​​(M is the number of data points in a frame). Then, the amplitude of each complex frequency value is calculated to obtain an amplitude spectrum of dimension N*M. According to the corresponding unvoiced signal and voiced signal labels, the amplitude spectrum of the N frames is divided into two categories: unvoiced signal amplitude spectrum and voiced signal amplitude spectrum, thus obtaining the amplitude spectrum of each frame of unvoiced signal and the amplitude spectrum of each frame of voiced signal.

[0281] After obtaining the amplitude spectrum of each frame of unvoiced sound signal, the bandwidth of the unvoiced sound signal can be determined based on the frequency value corresponding to the amplitude spectrum of each frame of unvoiced sound signal; after obtaining the amplitude spectrum of each frame of voiced sound signal, the bandwidth of the voiced sound signal can be determined based on the frequency value corresponding to the amplitude spectrum of each frame of voiced sound signal.

[0282] By performing time-frequency transformation on the unvoiced and voiced signals, and determining the bandwidth of the unvoiced and voiced signals based on the amplitude spectrum, a relatively accurate bandwidth can be obtained.

[0283] In an exemplary embodiment, determining the bandwidth of the unvoiced signal based on the amplitude spectrum of each frame of the unvoiced signal includes: determining the average amplitude spectrum of the unvoiced signal as the unvoiced amplitude spectrum based on the amplitude spectrum of each frame of the unvoiced signal; determining the maximum amplitude value in the unvoiced amplitude spectrum as the maximum unvoiced amplitude value, and determining a preset proportion of the maximum unvoiced amplitude value as the unvoiced reference amplitude value; determining the highest frequency value in the unvoiced amplitude spectrum whose amplitude value is greater than the unvoiced reference amplitude value as the first unvoiced frequency value; determining the frequency value in the unvoiced amplitude spectrum whose amplitude decreases the fastest with frequency as the second unvoiced frequency value; and determining the minimum frequency value between the first unvoiced frequency value and the second unvoiced frequency value as the effective bandwidth of the unvoiced signal.

[0284] The preset ratio is a pre-set ratio value used to determine the bandwidth of the audio signal; for example, it can be e.-6 .

[0285] The amplitude spectrum of all frames of the disclaimer signal is averaged to obtain the average amplitude spectrum of the disclaimer signal. This average amplitude spectrum is used as the disclaimer amplitude spectrum, which can be denoted as spectrum 0. The maximum amplitude value in the disclaimer amplitude spectrum is determined as the maximum disclaimer amplitude value Max0, and a preset proportion of the maximum disclaimer amplitude value Max0 (which can be represented by R) is determined as the disclaimer reference amplitude value, i.e., the disclaimer reference amplitude value is R*Max0. The amplitude values ​​in the disclaimer amplitude spectrum that are greater than the disclaimer reference amplitude value are determined, and the highest frequency value f0 among the frequency values ​​corresponding to these amplitude values ​​is determined. The highest frequency value f0 is used as the first disclaimer frequency value. The currently determined frequency value can be used as the current frequency value, and the frequency centered on the current frequency value is determined. The range corresponds to a first frequency value and a second frequency value. The first frequency value is less than the current frequency value, and the second frequency value is greater than the current frequency value. The first amplitude value corresponding to the first frequency value in the unvoiced amplitude spectrum is determined, and the second amplitude value corresponding to the second frequency value in the unvoiced amplitude spectrum is determined. The difference between the first amplitude value and the second amplitude value is determined as the amplitude drop value corresponding to the current frequency value. For each frequency value in the unvoiced amplitude spectrum, it is taken as the current frequency value and the corresponding amplitude drop value is determined. The frequency value corresponding to the largest amplitude drop value among the amplitude drop values ​​corresponding to each frequency value is determined as the second unvoiced frequency value f0. The first unvoiced frequency value f0 and the second unvoiced frequency value f0 are compared, and the minimum frequency value between the two is determined as the bandwidth of the unvoiced signal.

[0286] By determining the highest frequency value with an amplitude greater than the reference amplitude value of the unvoiced tone as the first unvoiced tone frequency value, determining the frequency value with the fastest amplitude decrease as the frequency as the second unvoiced tone frequency value, and determining the minimum frequency value between the first and second unvoiced tone frequency values ​​as the effective bandwidth of the unvoiced tone signal, the accuracy of determining the bandwidth of the unvoiced tone signal can be improved by comprehensively considering the amplitude value of the unvoiced tone signal and the change of the amplitude value with the frequency value.

[0287] In an exemplary embodiment, determining the bandwidth of the voiced signal based on the amplitude spectrum of the voiced signal in each frame includes:

[0288] Based on the amplitude spectrum of the voiced signal in each frame, the average amplitude spectrum of the voiced signal is determined as the voiced amplitude spectrum; the maximum amplitude value in the voiced amplitude spectrum is determined as the maximum voiced amplitude value, and a preset proportion of the maximum voiced amplitude value is determined as the voiced reference amplitude value; the highest frequency value in the voiced amplitude spectrum whose amplitude value is greater than the voiced reference amplitude value is determined as the first voiced frequency value; the frequency value in the voiced amplitude spectrum whose amplitude decreases the fastest with frequency is determined as the second voiced frequency value; the minimum frequency value between the first voiced frequency value and the second voiced frequency value is determined as the bandwidth of the voiced signal.

[0289] The preset ratio is a pre-set ratio value used to determine the bandwidth of the audio signal; for example, it can be e. -6 .

[0290] The amplitude spectrum of all frames of voiced signals is averaged to obtain the average amplitude spectrum of the voiced signal. This average amplitude spectrum is used as the voiced amplitude spectrum, which can be denoted as spectrum 1. The maximum amplitude value in the voiced amplitude spectrum is determined as the maximum voiced amplitude value Max1, and a preset proportion of the maximum voiced amplitude value Max1 (which can be represented by R) is determined as the voiced reference amplitude value, i.e., the voiced reference amplitude value is R*Max1. The amplitude values ​​in the voiced amplitude spectrum that are greater than the voiced reference amplitude value are determined, and the highest frequency value f1 among the frequency values ​​corresponding to these amplitude values ​​is determined. The highest frequency value f1 is used as the first voiced frequency value. The currently determined frequency value can be used as the current frequency value, and the frequency centered on the current frequency value is determined. The range corresponds to a first frequency value and a second frequency value, where the first frequency value is less than the current frequency value and the second frequency value is greater than the current frequency value. The first amplitude value corresponding to the first frequency value in the voiced amplitude spectrum is determined, and the second amplitude value corresponding to the second frequency value in the voiced amplitude spectrum is determined. The difference between the first amplitude value and the second amplitude value is determined as the amplitude decrease value corresponding to the current frequency value. For each frequency value in the voiced amplitude spectrum, it is taken as the current frequency value and its corresponding amplitude decrease value is determined. The frequency value corresponding to the largest amplitude decrease value among the amplitude decrease values ​​corresponding to each frequency value is determined as the second voiced frequency value ff1. The first voiced frequency value f1 and the second voiced frequency value ff1 are compared, and the minimum frequency value between the two is determined as the bandwidth of the voiced signal.

[0291] By determining the highest frequency value with an amplitude greater than the reference amplitude value of the voiced sound as the first voiced sound frequency value, the frequency value with the fastest amplitude decrease as the frequency as the second voiced sound frequency value, and the minimum frequency value between the first and second voiced sound frequency values ​​as the effective bandwidth of the voiced sound signal, the accuracy of determining the bandwidth of the voiced sound signal can be improved by comprehensively considering the amplitude value of the voiced sound signal and the change of the amplitude value with the frequency value.

[0292] In step S55, the bandwidth evaluation result of the audio to be processed is determined based on the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one of the second human voice segments.

[0293] The bandwidth of the unvoiced signal corresponding to at least one second voice segment can be statistically analyzed to obtain the bandwidth statistics of the unvoiced signal. Based on the relationship between the bandwidth statistics of the unvoiced signal and the bandwidth threshold, the bandwidth evaluation result of the unvoiced signal can be determined. The bandwidth of the voiced signal corresponding to at least one second voice segment can be statistically analyzed to obtain the bandwidth statistics of the voiced signal. Based on the relationship between the bandwidth statistics of the voiced signal and the bandwidth threshold, the bandwidth evaluation result of the voiced signal can be determined. The average value of the bandwidth evaluation results of the unvoiced signal and the bandwidth evaluation results of the voiced signal is determined as the bandwidth evaluation result of the audio to be processed.

[0294] Alternatively, the bandwidth evaluation results of the unvoiced signal and the voiced signal corresponding to each second voice segment can be determined separately, and then the bandwidth evaluation result of the second voice segment can be determined based on the bandwidth evaluation results of the unvoiced signal and the voiced signal, and then the bandwidth evaluation result of the audio to be processed can be determined based on the bandwidth evaluation results of at least one second voice segment.

[0295] In an exemplary embodiment, determining the bandwidth evaluation result of the audio to be processed based on the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one second voice segment includes: for each second voice segment, determining the bandwidth evaluation result of the unvoiced signal based on the bandwidth of the unvoiced signal, a first unvoiced bandwidth threshold, and a second unvoiced bandwidth threshold; for each second voice segment, determining the bandwidth evaluation result of the voiced signal based on the bandwidth of the voiced signal, a first voiced bandwidth threshold, and a second voiced bandwidth threshold; for each second voice segment, determining the average of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the voiced signal as the bandwidth evaluation result of the second voice segment; and determining the average of the bandwidth evaluation results of at least one second voice segment as the bandwidth evaluation result of the audio to be processed.

[0296] For each second voice segment, when determining the bandwidth evaluation result of the unvoiced signal, a first unvoiced bandwidth threshold and a second unvoiced bandwidth threshold can be preset. The first unvoiced bandwidth threshold is less than the second unvoiced bandwidth threshold. If the bandwidth of the unvoiced signal is less than or equal to the first unvoiced bandwidth threshold, the bandwidth evaluation result of the unvoiced signal is determined to be 0 (i.e., the bandwidth evaluation score of the unvoiced signal is 0 points); if the bandwidth of the unvoiced signal is greater than or equal to the second unvoiced bandwidth threshold, the bandwidth evaluation result of the unvoiced signal is determined to be full marks (i.e., the bandwidth evaluation score of the unvoiced signal is full marks); if the bandwidth of the unvoiced signal is greater than the first unvoiced bandwidth threshold and less than the second unvoiced bandwidth threshold, the bandwidth evaluation result of the unvoiced signal is determined according to the following formula:

[0297]

[0298] in, S represents the bandwidth evaluation result of the unvoiced signal. BW BW represents the maximum score for bandwidth evaluation. unvoiced BW represents the bandwidth of the unvoiced signal. unvoicedth1 BW represents the first unvoiced bandwidth threshold. unvoicedth2 This indicates the second unvoiced tone bandwidth threshold.

[0299] For each second voice segment, when determining the bandwidth evaluation result of the voiced signal, a first voiced bandwidth threshold and a second voiced bandwidth threshold can be preset. The first voiced bandwidth threshold is less than the second voiced bandwidth threshold. If the bandwidth of the voiced signal is less than or equal to the first voiced bandwidth threshold, the bandwidth evaluation result of the voiced signal is determined to be 0 (i.e., the bandwidth evaluation score of the voiced signal is 0 points); if the bandwidth of the voiced signal is greater than or equal to the second voiced bandwidth threshold, the bandwidth evaluation result of the voiced signal is determined to be full marks (i.e., the bandwidth evaluation score of the voiced signal is full marks); if the bandwidth of the voiced signal is greater than the first voiced bandwidth threshold and less than the second voiced bandwidth threshold, the bandwidth evaluation result of the voiced signal is determined according to the following formula:

[0300]

[0301] in, S represents the bandwidth evaluation result of the voiced signal. BW BW represents the maximum score for bandwidth evaluation. voiced BW represents the bandwidth of the voiced signal. voicedth1 BW represents the first voiced bandwidth threshold. voicedth2 This indicates the bandwidth threshold for the second voiced sound.

[0302] For each second voice segment, the average value of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the voiced signal is determined, and this average value is determined as the bandwidth evaluation result of the second voice segment; then the average value of the bandwidth evaluation results of at least one second voice segment is determined as the bandwidth evaluation result of the audio to be processed.

[0303] For each second voice segment, the bandwidth evaluation result of the unvoiced signal is determined based on the comparison between the bandwidth of the unvoiced signal and the first unvoiced bandwidth threshold and the second unvoiced bandwidth threshold. The bandwidth evaluation result of the voiced signal is determined based on the comparison between the bandwidth of the voiced signal and the first voiced bandwidth threshold and the second voiced bandwidth threshold. The average value of the bandwidth evaluation results of the unvoiced signals and the bandwidth evaluation results of the voiced quotation marks of all second voice segments is then used as the bandwidth evaluation result of the audio to be processed. This approach can yield a more accurate bandwidth evaluation result, thereby improving the accuracy of bandwidth evaluation.

[0304] By dividing the audio to be processed into multiple second time segments, identifying at least one second voice segment within these segments, and determining the bandwidth of the unvoiced signal and the bandwidth of the voiced signal in each second voice segment, and then statistically analyzing the bandwidth of the unvoiced signal and the bandwidth of the voiced signal in at least one second voice segment, the bandwidth evaluation result of the audio to be processed can be obtained, thus realizing the bandwidth evaluation of the audio to be processed.

[0305] In one exemplary embodiment, the recording quality factors include recording defects; Figure 6 This is a flowchart illustrating the recording quality evaluation of audio to be processed in response to recording defects, as described in this embodiment of the disclosure. Figure 6 As shown, the step of evaluating the recording quality of the audio to be processed according to at least one recording quality factor, and obtaining the evaluation result corresponding to each recording quality factor, includes:

[0306] In step S61, the audio to be processed is divided into multiple third time segments according to the third time length.

[0307] The third time length is a pre-set duration of a third time segment. The third time length can be longer than the first time length and the second time length.

[0308] The audio to be processed is divided according to the third time length, so that the waveform of the audio to be processed is divided into shorter time segments, resulting in multiple third time segments corresponding to the audio to be processed.

[0309] In step S62, the maximum absolute value of the amplitude of the audio to be processed is determined, and a first proportion of the maximum absolute value of the amplitude is determined as a first amplitude, and a second proportion of the maximum absolute value of the amplitude is determined as a second amplitude; the first proportion is less than the second proportion.

[0310] The absolute values ​​of the amplitudes of the audio to be processed are statistically analyzed to determine the maximum absolute amplitude value. A first proportion of this maximum absolute amplitude value is defined as the first amplitude, and a second proportion is defined as the second amplitude. Both the first and second proportions are proportionality coefficients less than 1; for example, the first proportion could be 0.8 and the second proportion could be 0.9.

[0311] In step S63, for each of the third time segments, the number of data points in the third time segment whose absolute amplitude value is between the first amplitude value and the second amplitude value is determined as the first data point count, and the number of data points in the third time segment whose absolute amplitude value is between the second amplitude value and the absolute value of the maximum amplitude value is determined as the second data point count.

[0312] For each third time segment, count the number of data points with absolute amplitude between the first and second amplitudes, and use this as the first data point count. Also count the number of data points with absolute amplitude between the second amplitude and the absolute value of the maximum amplitude, and use this as the second data point count.

[0313] In step S64, the third time segment where the number of the second data points is greater than the number of the first data points is determined as a recording defect segment.

[0314] Recording defects are primarily detected by detecting pops or crackles in the audio being processed. Pops typically have large absolute amplitude values. If there are a large number of data points with large absolute amplitude values, then pops or crackles are considered to exist, indicating a recording defect. The first and second data point counts are compared. If the second data point count is greater than the first, then the third time segment corresponding to both the second and first time point counts is identified as the recording defect segment.

[0315] In step S65, the recording defect evaluation result of the audio to be processed is determined based on the number of the recorded defective segments and the total number of the third time segments.

[0316] Determine the number of flawed recording segments in the audio to be processed, and determine the total number of third time segments in the audio to be processed. Based on the proportion of flawed recording segments in all third time segments, determine the evaluation result of the recording flaws in the audio to be processed.

[0317] In one exemplary embodiment, determining the recording defect evaluation result of the audio to be processed based on the number of the recording defective segments and the total number of the third time segments includes: determining the ratio of the number of the recording defective segments to the total number as the recording defect evaluation result of the audio to be processed.

[0318] The proportion of flawed recording segments in all third-time segments is determined, i.e., the ratio of the number of flawed recording segments to the total number of segments. This ratio is then used as the evaluation result of the recording defects in the audio to be processed. By determining the ratio of the number of flawed recording segments to the total number of segments as the evaluation result of the recording defects in the audio to be processed, a more accurate quantitative evaluation result of recording defects can be obtained.

[0319] In an exemplary embodiment, before determining the number of data points between the first amplitude and the second amplitude for each third time segment, the method further includes: determining a histogram of the third time segment according to the absolute value of the amplitude for each third time segment;

[0320] For each of the third time segments, determining the number of data points in the third time segment between the first amplitude and the second amplitude, as the first data point count, includes: determining the sum of the histogram between the first amplitude and the second amplitude, as the first data point count;

[0321] Determining the number of data points in the third time segment between the second amplitude and the absolute value of the maximum amplitude, as the second data point count, includes: determining the sum of the histogram between the second amplitude and the absolute value of the maximum amplitude, as the second data point count.

[0322] For each third time segment, count the number of data points corresponding to each absolute value of amplitude, and represent this count as a histogram. To determine the first number of data points, sum the histogram values ​​between the first and second amplitude values. To determine the second number of data points, sum the histogram values ​​between the second amplitude and the absolute value of the maximum amplitude value.

[0323] By determining the histogram of the third time segment according to the absolute value of the amplitude, and then determining the number of the first and second data points based on the histogram, data processing efficiency and evaluation efficiency of recording defects can be improved.

[0324] Figure 7 This is a block diagram illustrating an audio information evaluation device according to an exemplary embodiment. (Refer to...) Figure 7 The device includes a singing level evaluation module 71, a recording quality evaluation module 72, and a comprehensive evaluation module 73.

[0325] The singing level evaluation module 71 is configured to perform a singing level evaluation on the audio to be processed, and obtain the singing level evaluation result corresponding to the audio to be processed;

[0326] The recording quality evaluation module 72 is configured to perform a recording quality evaluation on the audio to be processed, and obtain the recording quality evaluation result corresponding to the audio to be processed;

[0327] The comprehensive evaluation module 73 is configured to determine the evaluation result of the audio to be processed based on the singing level evaluation result and the recording quality evaluation result.

[0328] Optionally, the recording quality evaluation module includes:

[0329] The recording quality factor evaluation unit is configured to perform recording quality evaluation on the audio to be processed according to at least one recording quality factor, and obtain the evaluation result corresponding to each recording quality factor.

[0330] The recording quality evaluation unit is configured to perform a weighted summation of the evaluation results corresponding to at least one recording quality factor based on the weight corresponding to each of the recording quality factors, to obtain the recording quality evaluation result.

[0331] Optionally, the recording quality factors include signal-to-noise ratio;

[0332] The recording quality factor evaluation unit includes:

[0333] The first segment division subunit is configured to divide the audio to be processed into multiple first time segments according to a first time length;

[0334] The first human voice segment determination subunit is configured to perform the determination of a first time segment containing human voice among a plurality of first time segments to obtain at least one first human voice segment.

[0335] The signal-to-noise ratio determination subunit is configured to perform, for each of the first human voice segments, determining the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment;

[0336] The signal-to-noise ratio (SNR) evaluation subunit is configured to perform an SNR evaluation result of the audio to be processed based on the SNR of at least one of the first human voice segments.

[0337] Optionally, the signal-to-noise ratio determination subunit is configured to perform:

[0338] For each of the first voice segments, blind source separation is performed on the first voice segment to obtain the voice audio signal in the first voice segment;

[0339] Based on the human voice audio signal, determine the human voice energy and noise energy in the first human voice segment;

[0340] The ratio of the human voice energy to the noise energy is determined to obtain the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment.

[0341] Optionally, the signal-to-noise ratio determination subunit is configured to perform:

[0342] The first time segments other than the first human voice segment are identified as noise segments among a plurality of the first time segments, to obtain at least one noise segment;

[0343] Determine the average energy of at least one of the noise segments as the noise energy;

[0344] For each of the first voice segments, the energy of the first voice segment is determined as the voice energy of the first voice segment;

[0345] The ratio of the human voice energy to the noise energy is determined to obtain the signal-to-noise ratio of the human voice audio signal to the noise audio signal in the first human voice segment.

[0346] Optionally, the signal-to-noise ratio evaluation subunit is configured to perform:

[0347] Determine the average signal-to-noise ratio of at least one of the first human voice segments to obtain the average signal-to-noise ratio;

[0348] The signal-to-noise ratio (SNR) evaluation result of the audio to be processed is determined based on the mean SNR, the first SNR threshold, and the second SNR threshold, wherein the first SNR threshold is less than the second SNR threshold.

[0349] Optionally, the recording quality factors include bandwidth;

[0350] The recording quality evaluation unit includes:

[0351] The second segment division subunit is configured to divide the audio to be processed into multiple second time segments according to a second time length;

[0352] The second voice segment determination subunit is configured to perform the determination of a second time segment containing a human voice among a plurality of second time segments, thereby obtaining at least one second voice segment;

[0353] The fundamental frequency detection subunit is configured to perform fundamental frequency detection on each frame of the second human voice segment, determine the frames in which the fundamental frequency is detected as voiced signals, and determine the frames in which the fundamental frequency cannot be detected as unvoiced signals;

[0354] The signal bandwidth determination subunit is configured to perform the determination of the bandwidth of the unvoiced signal in the second voice segment and the determination of the bandwidth of the voiced signal in the second voice segment;

[0355] The bandwidth evaluation subunit is configured to perform a bandwidth evaluation result of the audio to be processed based on the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one of the second vocal segments.

[0356] Optionally, the signal bandwidth determination subunit includes:

[0357] The time-frequency transformation submodule is configured to perform time-frequency transformation on each frame of unvoiced sound signal in the second voice segment to obtain the amplitude spectrum of each frame of unvoiced sound signal, and to perform time-frequency transformation on each frame of voiced sound signal in the second voice segment to obtain the amplitude spectrum of each frame of voiced sound signal.

[0358] The unvoiced bandwidth determination submodule is configured to determine the bandwidth of the unvoiced signal based on the amplitude spectrum of the unvoiced signal in each frame;

[0359] The voiced signal bandwidth determination submodule is configured to determine the bandwidth of the voiced signal based on the amplitude spectrum of the voiced signal in each frame.

[0360] Optionally, the unvoiced bandwidth determination submodule is configured to perform:

[0361] Based on the amplitude spectrum of the unvoiced signal in each frame, the average amplitude spectrum of the unvoiced signal is determined as the unvoiced amplitude spectrum;

[0362] The maximum amplitude value in the unvoiced tone amplitude spectrum is determined as the maximum amplitude value of the unvoiced tone, and a preset proportion of the maximum amplitude value of the unvoiced tone is determined as the reference amplitude value of the unvoiced tone.

[0363] The highest frequency value in the unvoiced tone amplitude spectrum whose amplitude value is greater than the unvoiced tone reference amplitude value is determined as the first unvoiced tone frequency value;

[0364] The frequency value at which the amplitude of the unvoiced tone decreases the fastest with frequency in the unvoiced tone amplitude spectrum is determined as the second unvoiced tone frequency value;

[0365] The frequency value that is the minimum of the first and second unvoiced frequency values ​​is determined as the bandwidth of the unvoiced signal;

[0366] The voiced bandwidth determination submodule is configured to execute:

[0367] Based on the amplitude spectrum of the voiced signal in each frame, the average amplitude spectrum of the voiced signal is determined as the voiced amplitude spectrum;

[0368] The maximum amplitude value in the voiced amplitude spectrum is determined as the maximum amplitude value of the voiced sound, and a preset proportion of the maximum amplitude value of the voiced sound is determined as the reference amplitude value of the voiced sound.

[0369] The highest frequency value in the voiced amplitude spectrum whose amplitude value is greater than the voiced reference amplitude value is determined as the first voiced frequency value;

[0370] The frequency value at which the amplitude of the voiced sound amplitude spectrum decreases the fastest with frequency is determined as the second voiced sound frequency value;

[0371] The frequency value that is the minimum of the first voiced frequency value and the second voiced frequency value is determined as the bandwidth of the voiced signal.

[0372] Optionally, the bandwidth evaluation subunit is configured to perform:

[0373] For each of the second human voice segments, the bandwidth evaluation result of the unvoiced signal is determined based on the bandwidth of the unvoiced signal, the first unvoiced bandwidth threshold, and the second unvoiced bandwidth threshold.

[0374] For each of the second human voice segments, the bandwidth evaluation result of the voiced signal is determined based on the bandwidth of the voiced signal, the first voiced bandwidth threshold, and the second voiced bandwidth threshold.

[0375] For each second voice segment, the average of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the voiced signal is determined as the bandwidth evaluation result of the second voice segment;

[0376] The average value of the bandwidth evaluation results of at least one of the second human voice segments is determined as the bandwidth evaluation result of the audio to be processed.

[0377] Optionally, the recording quality factors include recording defects;

[0378] The recording quality evaluation unit includes:

[0379] The third segment division subunit is configured to divide the audio to be processed into multiple third time segments according to a third time length;

[0380] An amplitude determination subunit is configured to perform the following operations: determine the maximum absolute amplitude value of the audio to be processed, determine a first proportion of the maximum absolute amplitude value as a first amplitude, and determine a second proportion of the maximum absolute amplitude value as a second amplitude; the first proportion is less than the second proportion.

[0381] The data point determination subunit is configured to perform, for each of the third time segments, determine the number of data points in the third time segment whose absolute amplitude value is between the first amplitude value and the second amplitude value, as the first data point count, and determine the number of data points in the third time segment whose absolute amplitude value is between the second amplitude value and the absolute value of the maximum amplitude value, as the second data point count;

[0382] The defective segment determination subunit is configured to determine a third time segment in which the number of the second data points is greater than the number of the first data points as a defective segment to be recorded;

[0383] The recording defect evaluation subunit is configured to determine the recording defect evaluation result of the audio to be processed based on the number of recording defect segments and the total number of the third time segments.

[0384] Optionally, the recording quality evaluation unit further includes:

[0385] The histogram-determining subunit is configured to perform a histogram determination for each of the third time segments based on the absolute value of the amplitude.

[0386] The data point determination subunit is configured to execute:

[0387] For each of the third time segments, the sum of the histogram between the first amplitude and the second amplitude is determined as the first data point count; the sum of the histogram between the second amplitude and the absolute value of the maximum amplitude is determined as the second data point count.

[0388] Optionally, the recording defect evaluation subunit is configured to perform:

[0389] The ratio of the number of defective recording segments to the total number is determined as the recording defect evaluation result of the audio to be processed.

[0390] Optionally, the recording quality evaluation module includes:

[0391] The device information acquisition unit is configured to acquire the device identifier and recording mode corresponding to the recording device of the audio to be processed;

[0392] The recording quality evaluation result determination unit is configured to acquire the hardware score corresponding to the device identifier and recording mode, and determine the hardware score as the recording quality evaluation result corresponding to the audio to be processed.

[0393] Optionally, the singing level evaluation module includes:

[0394] The fundamental frequency information acquisition unit is configured to acquire the fundamental frequency sequence of the audio to be processed and acquire the tone template corresponding to the audio to be processed;

[0395] The singing level evaluation unit is configured to perform a pitch evaluation result of the audio to be processed based on the fundamental frequency sequence and the pitch template, and use the pitch evaluation result as the singing level evaluation result.

[0396] Optionally, the elements in the fundamental frequency sequence include time points and fundamental frequency values, and the pitch template includes a start time, an end time, and a template frequency value corresponding to the note;

[0397] The singing level evaluation unit includes:

[0398] The time alignment subunit is configured to perform time alignment of the fundamental frequency sequence with the tone template based on the fundamental frequency value and the template frequency value, and update the time point corresponding to the fundamental frequency value to the alignment time point;

[0399] The frequency deviation determination subunit is configured to perform the following: determine the deviation between the fundamental frequency value and the template frequency value of each element based on the alignment time point.

[0400] The pitch evaluation subunit is configured to perform a pitch evaluation result of the audio to be processed based on the deviation.

[0401] Optionally, the frequency deviation determination subunit includes:

[0402] The frequency ratio determination submodule is configured to perform, for each of the notes, determine a target element between the start time and the end time for the alignment time, and determine the ratio of the fundamental frequency value corresponding to the target element to the template frequency value;

[0403] The frequency deviation determination submodule is configured to perform the operation of determining the deviation between the fundamental frequency value corresponding to the target element and the template frequency value based on the ratio.

[0404] Optionally, the frequency deviation determination submodule is configured to perform:

[0405] If the ratio is less than or equal to the first ratio threshold, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0406] If the ratio is greater than the first ratio threshold and the ratio is less than or equal to the value of the quadratic curve corresponding to the alignment time point of the target element, then it is determined that there is no deviation between the fundamental frequency value corresponding to the target element and the template frequency value; the quadratic curve is the curve between the start time and the end time, the values ​​of the quadratic curve corresponding to the start time and the end time are respectively the second ratio threshold, the value corresponding to the center time of the quadratic curve between the start time and the end time is the first ratio threshold, and the second ratio threshold is greater than the first ratio threshold;

[0407] If the ratio is greater than the value of the quadratic curve corresponding to the alignment time point of the target element, then the ratio is determined as the deviation between the fundamental frequency value corresponding to the target element and the template frequency value.

[0408] Optionally, the comprehensive evaluation module is configured to perform:

[0409] The evaluation results of the singing level and the recording quality are subjected to polynomial fitting to obtain the evaluation results of the audio to be processed.

[0410] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0411] Figure 8This is a block diagram illustrating an electronic device according to an exemplary embodiment. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0412] Reference Figure 8 The electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0413] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0414] Memory 804 is configured to store various types of data to support the operation of electronic device 800. Examples of this data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0415] Power supply component 806 provides power to various components of electronic device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 800.

[0416] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0417] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0418] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0419] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 may detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0420] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0421] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the audio information evaluation method described above.

[0422] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of an electronic device 800 to complete the aforementioned audio information evaluation method. Optionally, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0423] In an exemplary embodiment, a computer program product is also provided, including a computer program or computer instructions, which, when executed by a processor, implement the audio information evaluation method described above.

[0424] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0425] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An audio information evaluation method characterized by, The method comprises: evaluating a singing level of the to-be-processed audio to obtain a singing level evaluation result corresponding to the to-be-processed audio; evaluating a recording quality of the to-be-processed audio to obtain a recording quality evaluation result corresponding to the to-be-processed audio; determining an evaluation result of the to-be-processed audio according to the singing level evaluation result and the recording quality evaluation result; The method of evaluating the recording quality of the to-be-processed audio to obtain the recording quality evaluation result corresponding to the to-be-processed audio comprises: evaluating the recording quality of the to-be-processed audio according to at least one recording quality factor to obtain an evaluation result corresponding to each recording quality factor; performing weighted summation on the evaluation results corresponding to the at least one recording quality factor according to weights corresponding to each recording quality factor to obtain the recording quality evaluation result; The recording quality factor comprises a recording defect. The method of evaluating the recording quality of the to-be-processed audio according to at least one recording quality factor to obtain an evaluation result corresponding to each recording quality factor comprises: dividing the to-be-processed audio into a plurality of third time segments according to a third time length; determining a maximum amplitude absolute value of the to-be-processed audio, determining a first proportion of the maximum amplitude absolute value as a first amplitude, and determining a second proportion of the maximum amplitude absolute value as a second amplitude; the first proportion is less than the second proportion; for each third time segment, determining a number of data points with an amplitude absolute value between the first amplitude and the second amplitude in the third time segment as a first data point number, and determining a number of data points with an amplitude absolute value between the second amplitude and the maximum amplitude absolute value in the third time segment as a second data point number; determining a third time segment with the second data point number greater than the first data point number as a recording defect segment; determining a recording defect evaluation result of the to-be-processed audio according to a number of the recording defect segments and a total number of the third time segments.

2. The method of claim 1, wherein, The recording quality factor comprises a signal-to-noise ratio. The method of evaluating the recording quality of the to-be-processed audio according to at least one recording quality factor to obtain an evaluation result corresponding to each recording quality factor comprises: dividing the to-be-processed audio into a plurality of first time segments according to a first time length; determining a first time segment with human voice in the plurality of first time segments to obtain at least one first human voice segment; for each first human voice segment, determining a signal-to-noise ratio of a human voice signal and a noise signal in the first human voice segment; determining a signal-to-noise ratio evaluation result of the to-be-processed audio according to the signal-to-noise ratios of the at least one first human voice segment.

3. The method of claim 2, wherein, The method of determining the signal-to-noise ratio evaluation result of the to-be-processed audio according to the signal-to-noise ratios of the at least one first human voice segment comprises: determining an average value of the signal-to-noise ratios of the at least one first human voice segment to obtain a signal-to-noise ratio average value; According to the signal-to-noise ratio mean value, a first signal-to-noise ratio threshold value and a second signal-to-noise ratio threshold value, a signal-to-noise ratio evaluation result of the audio to be processed is determined, and the first signal-to-noise ratio threshold value is less than the second signal-to-noise ratio threshold value.

4. The method of claim 1, wherein, The recording quality factors include a bandwidth; The recording quality evaluation of the audio to be processed according to at least one recording quality factor respectively to obtain an evaluation result corresponding to each recording quality factor includes: The audio to be processed is divided into a plurality of second time segments according to a second time length; At least one second vocal segment is obtained by determining a second time segment with human voice in the plurality of second time segments; For each second vocal segment, a fundamental frequency of the second vocal segment is detected frame by frame, a frame with a detected fundamental frequency is determined as a voiced signal, and a frame without a detected fundamental frequency is determined as an unvoiced signal; A bandwidth of the unvoiced signal in the second vocal segment is determined, and a bandwidth of the voiced signal in the second vocal segment is determined; A bandwidth evaluation result of the audio to be processed is determined according to the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one second vocal segment.

5. The method of claim 4, wherein, The bandwidth of the unvoiced signal in the second vocal segment is determined, and the bandwidth of the voiced signal in the second vocal segment is determined, including: Time-frequency transformation is performed on each frame of unvoiced signal in the second vocal segment to obtain an amplitude spectrum of each frame of unvoiced signal, and time-frequency transformation is performed on each frame of voiced signal in the second vocal segment to obtain an amplitude spectrum of each frame of voiced signal; The bandwidth of the unvoiced signal is determined according to the amplitude spectrum of each frame of unvoiced signal; The bandwidth of the voiced signal is determined according to the amplitude spectrum of each frame of voiced signal.

6. The method of claim 5, wherein, The bandwidth of the unvoiced signal is determined according to the amplitude spectrum of each frame of unvoiced signal, including: An average amplitude spectrum of the unvoiced signal is determined according to the amplitude spectrum of each frame of unvoiced signal as an unvoiced amplitude spectrum; A maximum amplitude value in the unvoiced amplitude spectrum is determined as an unvoiced maximum amplitude value, and a preset proportion of the unvoiced maximum amplitude value is determined as an unvoiced reference amplitude value; A highest frequency value in the unvoiced amplitude spectrum with an amplitude value greater than the unvoiced reference amplitude value is determined as a first unvoiced frequency value; A frequency value with the fastest amplitude drop in the unvoiced amplitude spectrum is determined as a second unvoiced frequency value; The minimum frequency value of the first unvoiced frequency value and the second unvoiced frequency value is determined as the bandwidth of the unvoiced signal. The bandwidth of the voiced signal is determined according to the amplitude spectrum of each frame of voiced signal, including: An average amplitude spectrum of the voiced signal is determined according to the amplitude spectrum of each frame of voiced signal as a voiced amplitude spectrum; A maximum amplitude value in the voiced amplitude spectrum is determined as a voiced maximum amplitude value, and a preset proportion of the voiced maximum amplitude value is determined as a voiced reference amplitude value; A highest frequency value in the voiced amplitude spectrum with an amplitude value greater than the voiced reference amplitude value is determined as a first voiced frequency value; A frequency value with the fastest amplitude drop in the voiced amplitude spectrum is determined as a second voiced frequency value; determining a minimum value of the first voiced frequency value and the second voiced frequency value as a bandwidth of the voiced signal.

7. The method of claim 4, wherein, The bandwidth evaluation result of the to-be-processed audio is determined according to the bandwidth of the unvoiced signal and the bandwidth of the voiced signal corresponding to at least one of the second vocal fragments. For each of the second vocal fragments, a bandwidth evaluation result of the unvoiced signal is determined according to the bandwidth of the unvoiced signal, a first unvoiced bandwidth threshold and a second unvoiced bandwidth threshold. For each of the second vocal fragments, a bandwidth evaluation result of the unvoiced signal is determined according to the bandwidth of the unvoiced signal, a first unvoiced bandwidth threshold and a second unvoiced bandwidth threshold. For each of the second vocal fragments, an average value of the bandwidth evaluation result of the unvoiced signal and the bandwidth evaluation result of the unvoiced signal is determined as a bandwidth evaluation result of the second vocal fragment. An average value of the bandwidth evaluation results of at least one of the second vocal fragments is determined as the bandwidth evaluation result of the to-be-processed audio.

8. The method of claim 1, wherein, The recording flaw evaluation result of the to-be-processed audio is determined according to the number of the recording flaw fragments and the total number of the third time fragments. A ratio of the number of the recording flaw fragments to the total number is determined as the recording flaw evaluation result of the to-be-processed audio.

9. The method of claim 1, wherein, The singing level evaluation of the to-be-processed audio includes: A fundamental frequency sequence of the to-be-processed audio is obtained, and a pitch template corresponding to the to-be-processed audio is obtained. The singing level evaluation of the to-be-processed audio includes:

10. The method of claim 9, wherein, An element in the fundamental frequency sequence includes a time point and a fundamental frequency value, and the pitch template includes a start time, an end time and a template frequency value corresponding to a note. The singing level evaluation of the to-be-processed audio includes: The fundamental frequency sequence is time-aligned with the pitch template according to the fundamental frequency value and the template frequency value, and the time point corresponding to the fundamental frequency value is updated as an aligned time point. A deviation between the fundamental frequency value and the template frequency value of each of the elements is determined according to the aligned time point. The singing level evaluation of the to-be-processed audio is determined according to the deviation.

11. An audio information evaluation apparatus characterized by comprising: The singing level evaluation of the to-be-processed audio includes: a singing level evaluation module configured to perform singing level evaluation of the to-be-processed audio to obtain a singing level evaluation result corresponding to the to-be-processed audio; a recording quality evaluation module configured to perform recording quality evaluation of the to-be-processed audio to obtain a recording quality evaluation result corresponding to the to-be-processed audio; a comprehensive evaluation module configured to determine an evaluation result of the to-be-processed audio according to the singing level evaluation result and the recording quality evaluation result. The recording quality evaluation module includes: a recording quality factor evaluation unit configured to perform recording quality evaluation of the to-be-processed audio according to at least one recording quality factor to obtain an evaluation result corresponding to each of the recording quality factors. The recording quality evaluation unit is configured to perform weighted summation on evaluation results corresponding to the at least one recording quality factor according to weights corresponding to each of the recording quality factors to obtain the recording quality evaluation result. The recording quality factors include recording defects; The recording quality evaluation unit includes: A third segment division subunit configured to divide the audio to be processed into a plurality of third time segments according to a third time length; An amplitude determination subunit configured to determine a maximum amplitude absolute value of the audio to be processed, determine a first proportion of the maximum amplitude absolute value as a first amplitude, and determine a second proportion of the maximum amplitude absolute value as a second amplitude; the first proportion is less than the second proportion; A data point number determination subunit configured to determine, for each of the third time segments, a number of data points with amplitude absolute values between the first amplitude and the second amplitude in the third time segment as a first data point number, and determine a number of data points with amplitude absolute values between the second amplitude and the maximum amplitude absolute value in the third time segment as a second data point number; A defect segment determination subunit configured to determine third time segments with the second data point number greater than the first data point number as recording defect segments; A recording defect evaluation subunit configured to determine a recording defect evaluation result of the audio to be processed according to a number of the recording defect segments and a total number of the third time segments.

12. An electronic device, comprising: It includes: A processor; A memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the audio information evaluation method of any one of claims 1-10.

13. A computer-readable storage medium, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the audio information evaluation method of any one of claims 1-10.

Images