CN111182435B

CN111182435B - Testing method and device of voice equipment

Info

Publication number: CN111182435B
Application number: CN202010032773.3A
Authority: CN
Inventors: 刘孟美; 张雨晴
Original assignee: Beijing Xiaomi Pinecone Electronic Co Ltd
Current assignee: Beijing Xiaomi Pinecone Electronic Co Ltd
Priority date: 2020-01-13
Filing date: 2020-01-13
Publication date: 2021-11-09
Anticipated expiration: 2040-01-13
Also published as: CN111182435A

Abstract

The disclosure provides a testing method and device of voice equipment. The speech device includes a speaker and a microphone array, the method comprising: receiving an audio signal, the audio signal comprising at least one of: the method comprises the steps of carrying out parameter calculation on the basis of audio signals to obtain equipment performance parameters, determining the performance of voice equipment according to the equipment performance parameters, and outputting a performance test result, so as to obtain an accurate performance test result.

Description

Testing method and device of voice equipment

Technical Field

The present disclosure relates to the field of computer communications technologies, and in particular, to a method and an apparatus for testing a voice device.

Background

The voice control equipment such as an intelligent sound box is provided with a loudspeaker and a microphone array, wherein the loudspeaker is a transducer for converting an electric signal into an acoustic signal and is used for generating sound, and the microphone array is formed by arranging a certain number of microphones according to a certain rule and is used for sampling and processing the spatial characteristics of a sound field.

Before the voice control equipment leaves a factory or is applied, the performance of the voice control equipment is tested, and in the related art, some performance parameters of a microphone array are tested. However, the tested performance parameters are of a small variety, which results in an inaccurate performance test result of the voice control device.

Disclosure of Invention

In order to overcome the problems in the related art, the present disclosure provides a method and an apparatus for testing a voice device.

According to a first aspect of embodiments of the present disclosure, there is provided a method for testing a speech device, the speech device including a speaker and a microphone array, the method including:

receiving an audio signal, the audio signal comprising at least one of: audio signals collected by the microphone array when the test audio is played by the speaker, audio signals collected by a standard microphone when the test audio is played by the speaker, audio signals collected by the microphone array when the test audio is played by a high fidelity speaker;

performing parameter calculation based on the audio signal to obtain equipment performance parameters;

and determining the performance of the voice equipment according to the equipment performance parameters, and outputting a performance test result.

Optionally, the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the performing parameter calculation based on the audio signal to obtain the device performance parameter includes:

cutting the audio signals collected by each microphone to obtain single-frequency signals of the audio signals at different voltage levels;

for a single-frequency signal at each voltage level, determining the sensitivity of a pickup channel for transmitting the audio signal at the voltage level according to the single-frequency signal, wherein the pickup channel is a channel formed between a microphone for collecting the audio signal and the loudspeaker;

and determining a sensitivity level curve of the microphone according to the sensitivities of the microphone under all voltage levels corresponding to the microphone.

Optionally, the determining, for a single-frequency signal at each voltage level, a sensitivity of a pickup channel transmitting the audio signal at the voltage level according to the single-frequency signal includes:

for a single-frequency signal at each voltage level, determining a voltage of the single-frequency signal;

calculating the ratio of the voltage of the single-frequency signal to a target sound pressure to obtain the sensitivity of the pickup channel under the voltage level, wherein the target sound pressure comprises: and sound pressure corresponding to the preset volume.

Optionally, the test audio is a single-frequency sinusoidal signal sequence, the voltage is sequentially reduced from an initial voltage level to a preset voltage level by a preset tolerance, and every two adjacent groups of signals are separated by a mute signal; the ratio of the voltage of the single-frequency signal to the target sound pressure is calculated to obtain the sensitivity of the pickup channel under the voltage level, and the method comprises the following steps:

and calculating the ratio of the voltage of the single-frequency signal to the target sound pressure to obtain the sensitivity of the pickup channel under the current voltage level.

cutting the audio signal collected by each microphone to obtain a sweep frequency signal included in the audio signal;

and obtaining a sweep frequency curve of the voice equipment according to all the sweep frequency signals.

Optionally, the determining the performance of the voice device according to the device performance parameter and outputting a performance test result includes:

and determining whether the echo path of the voice equipment meets a preset condition or not according to the sweep frequency curve of the voice equipment, and outputting a determination result.

Optionally, the audio signal comprises: the method comprises the steps that a first audio signal collected by a microphone array when a test audio is played by a loudspeaker according to a preset volume, a second audio signal collected by a standard microphone when the test audio is played by the loudspeaker according to the preset volume, a third audio signal collected by the microphone array when a high-fidelity loudspeaker plays the test audio according to the same sound pressure level, and a fourth audio signal collected by the standard microphone when the test audio is played by the high-fidelity loudspeaker according to the same sound pressure level;

the performing parameter calculation based on the audio signal to obtain the device performance parameter includes:

cutting the first audio signal to obtain a first frequency sweeping signal, cutting the second audio signal to obtain a second frequency sweeping signal, cutting the third audio signal to obtain a third frequency sweeping signal, and cutting the fourth audio signal to obtain a fourth frequency sweeping signal;

obtaining a first frequency scanning curve based on the first frequency scanning signal, obtaining a second frequency scanning curve based on the second frequency scanning signal, obtaining a third frequency scanning curve based on the third frequency scanning signal, and obtaining a fourth frequency scanning curve based on the fourth frequency scanning signal;

calculating a difference value between the fourth frequency sweep curve and the second frequency sweep curve to obtain a vibration parameter of the loudspeaker;

calculating a difference value between the third frequency sweep curve and the first frequency sweep curve to obtain a total vibration parameter of the loudspeaker and the microphone array;

and calculating the difference value of the total vibration parameters and the vibration parameters of the loudspeaker to obtain the vibration parameters of the microphone array.

Optionally, the method further comprises:

acquiring equipment performance parameters of a plurality of voice equipment;

comparing the equipment performance parameters of the plurality of voice equipment and outputting a comparison result; alternatively, the first and second electrodes may be,

and directly outputting the device performance parameters of the plurality of voice devices.

According to a second aspect of embodiments of the present disclosure, there is provided a test system, comprising: the device comprises a test preparation module, a test module and a test result output module;

the test preparation module includes: the test environment calibration sub-module, the test parameter configuration sub-module and the audio uploading sub-module;

the testing module is configured to perform the testing method of any one of the first aspect.

According to a third aspect of the embodiments of the present disclosure, there is provided a testing apparatus of a voice device including a speaker and a microphone array, the apparatus including:

a receiving module configured to receive an audio signal, the audio signal comprising at least one of: audio signals collected by the microphone array when the test audio is played by the speaker, audio signals collected by a standard microphone when the test audio is played by the speaker, audio signals collected by the microphone array when the test audio is played by a high fidelity speaker;

a calculation module configured to perform parameter calculation based on the audio signal to obtain a device performance parameter;

and the determining module is configured to determine the performance of the voice equipment according to the equipment performance parameters and output a performance test result.

Optionally, the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the calculation module comprises:

the first cutting submodule is configured to cut the audio signals collected by each microphone to obtain single-frequency signals at different voltage levels included in the audio signals;

a first determination submodule configured to determine, for a single-frequency signal at each voltage level, a sensitivity at the voltage level of a pickup channel that transmits the audio signal, the pickup channel being a channel formed between a microphone that collects the audio signal and the speaker, from the single-frequency signal;

and the second determining submodule is configured to determine a sensitivity level curve of the microphone according to the sensitivities of the microphone at all voltage levels corresponding to the microphone.

Optionally, the first determining sub-module includes:

a determination unit configured to determine, for a single-frequency signal at each voltage level, a voltage of the single-frequency signal;

a calculating unit configured to calculate a ratio of a voltage of the single frequency signal to a target sound pressure, and obtain a sensitivity of the sound pickup channel at the voltage level, where the target sound pressure includes: and sound pressure corresponding to the preset volume.

Optionally, the test audio is a single-frequency sinusoidal signal sequence, the voltage is sequentially reduced from an initial voltage level to a preset voltage level by a preset tolerance, and every two adjacent groups of signals are separated by a mute signal;

the calculation unit is configured to calculate a ratio of the voltage of the single-frequency signal to the target sound pressure, and obtain the sensitivity of the sound pickup channel at the current voltage level.

a second cutting sub-module configured to cut the audio signal collected by each microphone to obtain a sweep frequency signal included in the audio signal;

and the first obtaining submodule is configured to obtain a frequency sweep curve of the voice equipment according to all frequency sweep signals.

Optionally, the determining module is configured to determine whether an echo path of the voice device meets a preset condition according to a sweep frequency curve of the voice device, and output a determination result.

the calculation module comprises:

a third cutting submodule configured to cut the first audio signal to obtain a first frequency sweeping signal, cut the second audio signal to obtain a second frequency sweeping signal, cut the third audio signal to obtain a third frequency sweeping signal, and cut the fourth audio signal to obtain a fourth frequency sweeping signal;

the second obtaining submodule is configured to obtain a first frequency scanning curve based on the first frequency scanning signal, obtain a second frequency scanning curve based on the second frequency scanning signal, obtain a third frequency scanning curve based on the third frequency scanning signal, and obtain a fourth frequency scanning curve based on the fourth frequency scanning signal;

the first calculation submodule is configured to calculate a difference value between the fourth frequency sweep curve and the second frequency sweep curve to obtain a vibration parameter of the loudspeaker;

a second calculation submodule configured to calculate a difference between the third frequency sweep curve and the first frequency sweep curve to obtain a total vibration parameter of the speaker and the microphone array;

a third calculation submodule configured to calculate a difference between the total vibration parameter and the vibration parameter of the speaker, resulting in a vibration parameter of the microphone array.

Optionally, the apparatus further comprises:

an acquisition module configured to acquire device performance parameters of a plurality of voice devices;

the comparison module is configured to compare the equipment performance parameters of the plurality of voice equipment and output a comparison result; alternatively, the first and second electrodes may be,

an output module configured to directly output device performance parameters of the plurality of speech devices.

According to a fourth aspect of embodiments of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any one of the above first aspects.

According to a fifth aspect of the embodiments of the present disclosure, there is provided a testing apparatus of a speech device, including:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to:

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:

in an embodiment of the present disclosure, in testing audio performance of a speech device, an audio signal is received, where the audio signal includes at least one of: the method comprises the steps of carrying out parameter calculation on the basis of audio signals to obtain equipment performance parameters, determining the performance of voice equipment according to the equipment performance parameters, and outputting a performance test result, so as to obtain an accurate performance test result.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

FIG. 1 is a flow chart illustrating a method for testing a speech device according to an exemplary embodiment;

FIG. 2 is a flow chart illustrating another method of testing a speech device according to an exemplary embodiment;

FIG. 3 is a schematic diagram illustrating a configuration interface of an online test platform in accordance with an exemplary embodiment;

FIG. 4 is a schematic diagram illustrating a configuration interface in an audio format in accordance with an illustrative embodiment;

FIG. 5 is a schematic diagram illustrating a configuration interface for a parameter to be tested in accordance with an exemplary embodiment;

FIG. 6 illustrates a test plot of a frequency sweep curve according to an exemplary embodiment;

FIG. 7 is a block diagram of a testing apparatus for a speech device, according to an example embodiment;

fig. 8 is a schematic diagram illustrating a structure of an apparatus for testing a speech device according to an exemplary embodiment.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

Fig. 1 is a flowchart illustrating a method for testing a voice device according to an exemplary embodiment, where the method illustrated in fig. 1 is applied to a test system, and the test system may be an online test system or an offline test system, and the method includes:

in step 101, an audio signal is received, the audio signal comprising at least one of: the audio signal collected by the microphone array when the test audio is played by the loudspeaker, the audio signal collected by the standard microphone when the test audio is played by the loudspeaker, and the audio signal collected by the microphone array when the test audio is played by the high-fidelity loudspeaker.

The voice device comprises a loudspeaker and a microphone array, and the voice device can be an intelligent voice device, such as an intelligent sound box.

In a scene of testing a loudspeaker in a voice device, the loudspeaker is used for playing test audio according to a preset volume, a standard microphone is used for collecting audio signals, and the standard microphone sends the collected audio signals to an online test platform or an offline test system. The preset volume may be the maximum volume or half of the maximum volume, etc.

In a scene of testing a microphone array in the voice equipment, a high-fidelity loudspeaker is used for playing test audio according to preset volume, the microphone array in the voice equipment is used for collecting audio signals, and the microphone array in the voice equipment sends the collected audio signals to an online test system or an offline test system.

In a scene of performing echo test on the voice equipment, a loudspeaker in the voice equipment is used for playing test audio according to preset volume, a microphone array in the voice equipment is used for collecting audio signals, and the microphone array sends the collected audio signals to an online test system or an offline test system. The preset volume may be the maximum volume or half of the maximum volume, etc.

The test audio used in different test scenes can be the same or different, and can be set according to needs or experience.

In step 102, a parameter calculation is performed based on the audio signal to obtain a device performance parameter.

In the scenario of testing a loudspeaker, the audio signal comprises: the method comprises the steps that a loudspeaker plays audio signals collected by a standard microphone when testing audio according to preset volume, and a testing system processes the audio signals to obtain performance parameters of the loudspeaker, wherein the performance parameters comprise harmonic distortion parameters, bottom noise parameters, a sweep frequency curve of the loudspeaker, a maximum sound pressure level, abnormal sound vibration sound parameters, sensitivity level curves under different voltages and harmonic distortion curves under different voltages. The basic performance of the loudspeaker can be judged according to the performance parameters of the loudspeaker.

In the scenario of testing the microphone array, the audio signal comprises: the method comprises the steps that when a high-fidelity loudspeaker plays test audio, audio signals collected by a microphone array are processed by a test system, and performance parameters of the microphone array, including sensitivity, harmonic distortion parameters, signal-to-noise ratio, bottom noise parameters, a sweep frequency curve of the microphone array and an external sealing type curve of the microphone, and a sensitivity level curve and a harmonic distortion curve under different sound pressure levels, are obtained. The basic performance of the microphone array can be judged according to the performance parameters of the microphone array.

In a scenario of performing an echo test on a speech device, an audio signal includes: the method comprises the steps that a loudspeaker in the voice equipment plays an audio signal collected by each microphone in a microphone array when testing audio according to a preset volume, the preset volume can be the maximum volume or half of the maximum volume, and the like, so that performance parameters under an Echo scene, a frequency sweep curve under the Echo situation, a convergence curve and an ERLE (Echo Return Loss gain) curve of a classical Echo cancellation algorithm, a sensitivity level curve and a harmonic distortion curve under different voltages are obtained.

A sound pickup channel is formed between each microphone and the loudspeaker in the microphone array, the loudspeaker plays sound, the microphone receives sound, and the audio received by the microphone is the audio of the sound pickup channel.

Typically, the audio signal comprises: various types of signals such as a frequency sweep signal, a single frequency signal, a white noise signal, and the like. The device performance parameter may include a sensitivity level curve of the speech device, fig. 2 shows a flowchart of another testing method for the speech device according to an exemplary embodiment, referring to fig. 2, the testing system may determine the sensitivity level curve of the speech device by, in step 1021, cutting the audio signal for the audio signal collected by each microphone to obtain a single-frequency signal at different voltage levels included in the audio signal; in step 1022, for a single-frequency signal at each voltage level, determining, according to the single-frequency signal, a sensitivity of a sound pickup channel for transmitting an audio signal at the voltage level, where the sound pickup channel is a channel formed between a microphone and a speaker for collecting the audio signal; in step 1023, a sensitivity level curve for the microphone is determined based on the sensitivity at all voltage levels for the microphone.

For step 1021, for example, a VAD (Voice Activity Detection) endpoint Detection method may be used to cut the audio signal, so as to implement automatic cutting of the audio signal. And judging the starting point by using a VAD method, and automatically cutting signals such as single frequency, white noise, frequency sweep signals and the like required by the test from the received audio signals according to the starting point and the preset time setting for playing the test audio.

Other suitable cutting methods may also be used, and embodiments of the present disclosure are not limited.

For step 1022, for a single-frequency signal at each voltage level, an operation of determining, according to the single-frequency signal, a sensitivity of a sound pickup channel for transmitting an audio signal at the voltage level may be implemented as follows: for the single-frequency signal under each voltage level, determine the voltage of this single-frequency signal, calculate the ratio of the voltage of this single-frequency signal and target acoustic pressure, obtain the sensitivity of this pickup channel under this voltage level, wherein, target acoustic pressure includes: and a loudspeaker in the voice equipment plays the sound pressure of the test audio according to the preset volume.

The test audio frequency is a single-frequency sinusoidal signal sequence, the voltage is sequentially reduced from the initial voltage level to the preset voltage level by the preset tolerance, every two adjacent groups of signals are separated by the mute signal, and the test system can calculate the ratio of the voltage of the single-frequency signal to the target sound pressure to obtain the sensitivity of the pickup channel under the current voltage level.

Exemplarily, the test audio is a 1kHz single-frequency sinusoidal signal sequence, the voltage is sequentially reduced from 0dB to a preset voltage level by a preset tolerance of 6dB, and the break time between each two adjacent groups of signals is a 2s mute sequence, after the test system receives a 1kHz single-frequency signal at a certain voltage level, the ratio of the voltage of the single-frequency signal to the target sound pressure is calculated, so as to obtain the sensitivity of the pickup channel at the current voltage level.

For step 1023, the product of the logarithm of the sensitivity of the sound pickup channel at different voltage levels and 20 may be calculated for the sound pickup channel formed by each microphone, taking 10 as a base, to obtain the sensitivity level of the sound pickup channel. And obtaining a sensitivity level curve of the microphone according to the sensitivity levels of the pickup channel under different voltage levels.

The device performance parameters may include a sweep curve of the speech device. Under the scene of carrying out echo test on the voice equipment, the test system receives audio signals transmitted by each microphone through a respective pickup channel, cuts each audio signal to obtain frequency sweep signals included by each audio signal, and obtains a frequency sweep curve according to each frequency sweep signal.

In the subsequent operation, the test system can determine whether the echo path of the voice equipment meets the preset condition according to all the sweep frequency curves, and outputs a determination result. Based on the determination result, the influence of the echo path on the microphone parameters and the loudspeaker parameters in the echo test scene is obtained, and the unreasonable structural design of the voice equipment is determined. The preset conditions may be set according to needs and experience, for example, the preset conditions include: the sweep frequency curve has no obvious defect, or the sensitivity level curve is linear and the like.

The device performance parameters may include a vibration parameter of the speaker and a vibration parameter of the microphone array. In a scenario of testing vibration parameters of a speaker and a microphone array in a speech device, an audio signal includes: the method comprises the steps of collecting a first audio signal by a microphone array when a loudspeaker plays a test audio according to a preset volume, collecting a second audio signal by a standard microphone when the loudspeaker plays the test audio according to the preset volume, collecting a third audio signal by the microphone array when a high-fidelity loudspeaker plays the test audio according to the same sound pressure level, and collecting a fourth audio signal by the standard microphone when the high-fidelity loudspeaker plays the test audio according to the same sound pressure level. The preset volume may be the maximum volume or half of the maximum volume, etc.

The test system cuts the first audio signal to obtain a first frequency sweeping signal, cuts the second audio signal to obtain a second frequency sweeping signal, cuts the third audio signal to obtain a third frequency sweeping signal, and cuts the fourth audio signal to obtain a fourth frequency sweeping signal.

After the signal cutting is finished, the test system obtains a first frequency scanning curve based on the first frequency scanning signal, obtains a second frequency scanning curve based on the second frequency scanning signal, obtains a third frequency scanning curve based on the third frequency scanning signal, and obtains a fourth frequency scanning curve based on the fourth frequency scanning signal. Obtaining a frequency sweep curve based on a frequency sweep signal is prior art, and details are not described herein in the embodiments of the present disclosure.

The testing system calculates a difference value between the fourth frequency sweep curve and the second frequency sweep curve to obtain a vibration parameter of the loudspeaker, the vibration parameter of the loudspeaker reflects the influence of vibration on the performance of the loudspeaker, the difference value between the third frequency sweep curve and the first frequency sweep curve is calculated to obtain a total vibration parameter of the loudspeaker and the microphone array, the total vibration parameter reflects the influence of vibration on the performance of the loudspeaker and the performance of the microphone array, the difference value between the total vibration parameter and the vibration parameter of the loudspeaker is calculated to obtain a vibration parameter of the microphone array, and the vibration parameter of the microphone array reflects the influence of vibration on the performance of the microphone array.

And calculating the difference value of the two sweep frequency curves, specifically calculating the difference value of the ordinate of the same abscissa in the two sweep frequency curves.

M represents a standard microphone, M _ D represents a fourth frequency sweep curve, M _ O represents a second frequency sweep curve, S represents a microphone array in the voice equipment, namely, a microphone to be tested, S _ D represents a third frequency sweep curve, and S _ O represents a first frequency sweep curve. Vibration parameters of the loudspeaker, namely Vib _ speaker, M _ D-M _ O; the total vibration parameter of the loudspeaker and microphone array Vib _ all is S _ D-S _ O. The vibration parameter Vib _ mic of the microphone array is (S _ D-S _ O) - (M _ D-M _ O).

The embodiment of the disclosure innovatively provides a method for measuring the influence of vibration on the performance of a loudspeaker and the performance of a microphone array, and based on the method, the comprehensive and accurate test on the performance of voice equipment is realized.

In step 103, the performance of the voice device is determined according to the device performance parameters, and a performance test result is output.

After obtaining the performance parameters of the equipment, the test system determines the performance of the voice equipment according to the performance parameters of the equipment and outputs a performance test result.

The performance of the voice equipment can be determined according to the equipment performance parameters by various means such as threshold comparison, standard curve comparison, determination of whether the test curve is complete and flawless, determination of whether the test curve is linear, and the like. The performance test results may include: test passed, test failed, failed parameters, etc.

In the embodiment of the disclosure, in the process of testing the audio performance of the voice device, the test parameters are added, including the sensitivity level curve and the sweep frequency curve of the voice device, various parameters of the speaker and the vibration parameter of the microphone, and by testing the parameters of the voice device, the audio performance of the voice device can be comprehensively analyzed, and an accurate performance test result can be obtained.

The embodiment of the disclosure provides a test system, which comprises a test preparation module, a test module and a test result output module; wherein, the test preparation module includes: the test environment calibration sub-module, the test parameter configuration sub-module and the audio uploading sub-module; the testing module is used for executing the testing method of the voice equipment provided by the disclosure.

The test system may be an online test system for online testing of audio performance of the speech device. And the test result output module is used for outputting information such as comparison between the test result of the voice equipment and a historical record.

The test system may further include: the system comprises a standard document module and a platform management module, wherein the standard document module is used for hardware information management, standard document management and the like, and the hardware information and the standard document introduce the whole test flow and the attention; the platform management module is used for user grouping management, system access control and the like, and is convenient for relevant project personnel to see the current project progress.

During the actual test process, the test system may perform the following operations in sequence: firstly, testing environment calibration, environment configuration and testing instrument configuration, and collecting audio signals by a Device Under Test (DUT); step two, a test directory is newly established and audio is uploaded; thirdly, configuring audio parameters and testing parameters; fourthly, cutting audio and calculating parameters; and fifthly, comparing results. The audio in the second step and the third step is the audio signal described previously in the embodiments of the present disclosure. The DUT is used for collecting audio signals for all subsequent data calculation, and the method has the characteristic of less audio signal collection amount.

Fig. 3 is a schematic diagram of a configuration interface of a test system according to an exemplary embodiment, and referring to fig. 3, in the process of creating a test set on the test system, parameters such as test time, cutting algorithm version, calculation algorithm version, test instruments, test environment, and the like need to be set. The calculation algorithm is an algorithm for obtaining a sweep frequency curve based on the sweep frequency signal.

After the test set is successfully established, the test system obtains the uploaded audio signal, the configuration parameters, specifically, the audio format and the parameters to be tested, fig. 4 is a schematic diagram of a configuration interface of an audio format according to an exemplary embodiment, fig. 5 is a schematic diagram of a configuration interface of a parameter to be tested according to an exemplary embodiment, see the types of parameters shown in fig. 4 and fig. 5.

The parameters in fig. 5 are explained below.

MIC query fields:

SENSITIVITY: sensitivity level (dB), microphone output voltage corresponding to a sound pressure level of 94 dBSPL. THD: harmonic distortion, frequency 1000Hz, 94dBSPL sound pressure level. THD _ sweet: harmonic distortion based on frequency sweep (94 dBSPL). NOISE: noise level, classified as high, medium, and low frequency cases. SNR: signal to noise ratio. DELAY: the delay differences between different channels are in the sequence of [1,2], [1,3] [1,4] [2,3] [2,4] [3,4] by taking 4 channels as an example. REF _ DELAY: time delay difference of each channel and the reference channel signal. CORR: the correlation between different channels is represented by the sequence of [1,2], [1,3] [1,4] [2,3] [2,4] [3,4] in the example of 4 channels. PHASE: phase curves for different channels. FR _ 94: frequency response curve based on frequency sweep (94 dBSPL). SEN _ THD: sensitivity level curve, harmonic distortion curve, 1000Hz case at different sound pressure levels. FR _ THD: frequency response curves and harmonic distortion are based on different sound pressure levels. SEALING _ OUT: and (4) external tightness. SEALING _ IN: internal tightness. AEC _ MAX: at maximum volume, AEC performance. AEC _ HALF: at half the volume, AEC performance. AEC _ SEN _ THD _ MAX: sensitivity value and harmonic distortion (maximum volume) under different electric signal amplitudes in the echo condition. AEC _ THD _ sweet _ MAX: the echo case is based on harmonic distortion (maximum volume) of the frequency sweep. AEC _ FR _ sweet _ MAX: the echo case is based on the frequency response curve (maximum volume) of the frequency sweep. AEC _ SEN _ THD _ HALF: sensitivity value and harmonic distortion (half volume) under different electric signal amplitudes in the echo case. AEC _ THD _ sweet _ HALF: the echo case is based on harmonic distortion (half the volume) of the frequency sweep. AEC _ FR _ sweet _ HALF: the echo case is based on the frequency response curve (half the volume) of the frequency sweep. Directivity: directionality. Aliasing: aliasing is performed. Resonance _ point: a resonance point. Clipping: an amplitude cut sound pressure level.

SPK query fields:

SENSITIVITY _ MAX: maximum volume sensitivity level, frequency 1000 Hz. SENSITIVITY _ HALF: half the volume sensitivity level, at a frequency of 1000 Hz. THD _ MAX: maximum volume harmonic distortion, frequency 1000 Hz. THD _ HALF: half the volume harmonic distortion, frequency 1000 Hz. THD _ sweet _ MAX: maximum volume harmonic distortion, frequency sweep. THD _ sweet _ HALF: half volume harmonic distortion, frequency sweep. NOISE: and (5) background noise. SNR: signal to noise ratio. FR _ MAX: maximum volume frequency response curve, frequency sweep. FR _ HALF: a half volume frequency response curve and frequency sweep. SEN _ THD: sensitivity level curve, harmonic distortion curve, 1000Hz case at different sound pressure levels. Rub & Buzz _ MAX: maximum volume Rub & Buzz. Rub & Buzz _ HALF: half volume Rub & Buzz. VIB _ MAX: maximum volume shock effect. VIB _ HALF: half the volume of the vibration effect. Max _ level: the maximum sound pressure level of the speaker. Rub _ p: the loudspeaker has abnormal sound and vibration sound.

Fig. 4 and 5 show a large number of parameters, and the test system can set a large number of parameters, and can flexibly control the value of each parameter, so as to maximize the audio utilization rate while ensuring the test efficiency.

The test system automatically cuts the audio signal using an audio cutting algorithm. Commonly referred to as VAD algorithms, audio cutting algorithms based on energy and short-term zero-crossing rates, etc. may be used.

And testing the audio performance of the voice equipment based on the early-stage setting to obtain a performance test result of the voice equipment.

The test system may store device performance parameters obtained for the voice device after testing the audio performance of the voice device. For the online test system, the device performance parameters of a plurality of voice devices can be obtained, the device performance parameters of the plurality of voice devices are compared, and a comparison result is output; or directly outputting the device performance parameters of a plurality of voice devices for viewing and comparison.

For example, the online test system may obtain frequency sweep curves of microphones at the same position in a microphone array in a plurality of voice devices, plot the frequency sweep curves of the microphones at the same position in a test chart, and output the test chart. Fig. 6 is a test graph of a sweep curve, where fig. 6 shows the sweep curve of a microphone at the same location in three speech devices, according to an exemplary embodiment.

By the method, the performance parameters of the voice devices are transversely measured on line, and comparison and query are facilitated.

The embodiment of the disclosure provides a new testing method of voice equipment, which creatively determines a plurality of parameters having great significance on performance evaluation of the voice equipment through multiple times of exploration and practice, and in the actual testing process, the audio performance of the voice equipment is comprehensively tested based on the testing parameters, the audio performance of the voice equipment is comprehensively analyzed, and an accurate performance testing result is obtained.

While, for purposes of simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present disclosure is not limited by the order of acts, as some steps may, in accordance with the present disclosure, occur in other orders and concurrently.

Further, those skilled in the art should also appreciate that the embodiments described in the specification are exemplary embodiments and that acts and modules referred to are not necessarily required by the disclosure.

Corresponding to the embodiment of the application function implementation method, the disclosure also provides an embodiment of an application function implementation device and a corresponding terminal.

Fig. 7 is a block diagram illustrating an apparatus for testing a speech device including a speaker and an array of microphones, according to an example embodiment, the apparatus comprising: a receiving module 21, a calculating module 22 and a determining module 23; wherein the content of the first and second substances,

the receiving module 21 is configured to receive an audio signal, the audio signal including at least one of: audio signals collected by the microphone array when the test audio is played by the speaker, audio signals collected by a standard microphone when the test audio is played by the speaker, audio signals collected by the microphone array when the test audio is played by a high fidelity speaker;

the calculation module 22 is configured to perform parameter calculation based on the audio signal, so as to obtain a device performance parameter;

the determining module 23 is configured to determine the performance of the voice device according to the device performance parameter, and output a performance test result.

In an alternative embodiment, on the basis of the testing apparatus of the speech device shown in fig. 7, the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the calculation module 22 may include: the device comprises a first cutting submodule, a first determining submodule and a second determining submodule; wherein the content of the first and second substances,

the first cutting submodule is configured to cut the audio signals collected by each microphone to obtain that the audio signals comprise single-frequency signals at different voltage levels;

the first determining submodule is configured to determine, for a single-frequency signal at each voltage level, sensitivity, at the voltage level, of a pickup channel that transmits the audio signal, the pickup channel being a channel formed between a microphone that collects the audio signal and the speaker, according to the single-frequency signal;

the second determining submodule is configured to determine a sensitivity level curve of the microphone according to the sensitivities of the microphone at all voltage levels corresponding to the microphone.

In an optional embodiment, the first determining sub-module may include: a determination unit and a calculation unit; wherein the content of the first and second substances,

the determination unit is configured to determine, for a single-frequency signal at each voltage level, a voltage of the single-frequency signal;

the calculation unit is configured to calculate a ratio of a voltage of the single-frequency signal to a target sound pressure, and obtain a sensitivity of the pickup channel at the voltage level, where the target sound pressure includes: and sound pressure corresponding to the preset volume.

In an optional embodiment, the test tone is a single-frequency sinusoidal signal sequence, the voltage is sequentially reduced from an initial voltage level to a preset voltage level by a preset tolerance, and every two adjacent groups of signals are separated by a mute signal;

the calculation unit may be configured to calculate a ratio of the voltage of the single-frequency signal to the target sound pressure, and obtain the sensitivity of the sound pickup channel at a current voltage level.

In an alternative embodiment, on the basis of the testing apparatus of the speech device shown in fig. 7, the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the calculation module 22 may include: a second cutting submodule and a first obtaining submodule; wherein the content of the first and second substances,

the second cutting submodule is configured to cut the audio signal acquired by each microphone to obtain a sweep frequency signal included in the audio signal;

the first obtaining submodule is configured to obtain a frequency sweep curve of the voice device according to all frequency sweep signals.

In an optional embodiment, the determining module may be configured to determine whether an echo path of the speech device satisfies a preset condition according to a sweep frequency curve of the speech device, and output a determination result.

In an alternative embodiment, on the basis of the testing apparatus of the speech device shown in fig. 7, the audio signal comprises: the method comprises the steps that a first audio signal collected by a microphone array when a test audio is played by a loudspeaker according to a preset volume, a second audio signal collected by a standard microphone when the test audio is played by the loudspeaker according to the preset volume, a third audio signal collected by the microphone array when a high-fidelity loudspeaker plays the test audio according to the same sound pressure level, and a fourth audio signal collected by the standard microphone when the test audio is played by the high-fidelity loudspeaker according to the same sound pressure level;

the calculation module 22 may include: the device comprises a third cutting submodule, a second obtaining submodule, a first calculating submodule, a second calculating submodule and a third calculating submodule; wherein the content of the first and second substances,

the third cutting submodule is configured to cut the first audio signal to obtain a first frequency sweeping signal, cut the second audio signal to obtain a second frequency sweeping signal, cut the third audio signal to obtain a third frequency sweeping signal, and cut the fourth audio signal to obtain a fourth frequency sweeping signal;

the second obtaining submodule is configured to obtain a first frequency sweep curve based on the first frequency sweep signal, obtain a second frequency sweep curve based on the second frequency sweep signal, obtain a third frequency sweep curve based on the third frequency sweep signal, and obtain a fourth frequency sweep curve based on the fourth frequency sweep signal;

the second calculation submodule is configured to calculate a difference value between the third frequency sweep curve and the first frequency sweep curve to obtain a total vibration parameter of the loudspeaker and the microphone array;

the third calculation sub-module is configured to calculate a difference value between the total vibration parameter and the vibration parameter of the loudspeaker to obtain the vibration parameter of the microphone array.

In an alternative embodiment, on the basis of the testing apparatus of the speech device shown in fig. 7, the apparatus may further include: the device comprises an acquisition module, a comparison module and an output module; wherein the content of the first and second substances,

the acquisition module is configured to acquire device performance parameters of a plurality of voice devices;

the comparison module is configured to compare the device performance parameters of the plurality of voice devices and output a comparison result; alternatively, the first and second electrodes may be,

the output module is configured to directly output the device performance parameters of the plurality of voice devices.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. One of ordinary skill in the art can understand and implement it without inventive effort.

In an exemplary embodiment, there is also provided an apparatus for testing a voice device, fig. 8 is a schematic structural diagram illustrating an apparatus for testing a voice device according to an exemplary embodiment, and the apparatus illustrated in fig. 8 may include: a memory 320, a processor 330, and an external interface 340 connected by an internal bus 310;

wherein the external interface 340 is configured to receive an audio signal, the audio signal including at least one of: audio signals collected by the microphone array when the test audio is played by the speaker, audio signals collected by a standard microphone when the test audio is played by the speaker, audio signals collected by the microphone array when the test audio is played by a high fidelity speaker;

a memory 320 for storing machine readable instructions corresponding to the scan;

a processor 330 configured to read the machine-readable instructions on the memory 320 and execute the instructions to:

In an exemplary embodiment, there is also provided a non-transitory computer readable storage medium, such as the memory 1604 comprising instructions that when executed by the processor 1620 of the apparatus 1600, enable the apparatus 1600 to perform a method of testing a speech device, the method comprising: receiving an audio signal, the audio signal comprising at least one of: audio signals collected by the microphone array when the test audio is played by the speaker, audio signals collected by a standard microphone when the test audio is played by the speaker, audio signals collected by the microphone array when the test audio is played by a high fidelity speaker; performing parameter calculation based on the audio signal to obtain equipment performance parameters; and determining the performance of the voice equipment according to the equipment performance parameters, and outputting a performance test result.

The non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method of testing a speech device, the speech device comprising a speaker and an array of microphones, the method comprising:

receiving an audio signal, the audio signal comprising: the method comprises the steps that a first audio signal collected by a microphone array when a loudspeaker plays a test audio according to a preset volume, a second audio signal collected by a standard microphone when the loudspeaker plays the test audio according to the preset volume, a third audio signal collected by the microphone array when a high-fidelity loudspeaker plays the test audio according to the same sound pressure level, and a fourth audio signal collected by the standard microphone when the high-fidelity loudspeaker plays the test audio according to the same sound pressure level;

performing parameter calculation based on the audio signal to obtain a device performance parameter, including: cutting the first audio signal to obtain a first frequency sweeping signal, cutting the second audio signal to obtain a second frequency sweeping signal, cutting the third audio signal to obtain a third frequency sweeping signal, and cutting the fourth audio signal to obtain a fourth frequency sweeping signal; obtaining a first frequency scanning curve based on the first frequency scanning signal, obtaining a second frequency scanning curve based on the second frequency scanning signal, obtaining a third frequency scanning curve based on the third frequency scanning signal, and obtaining a fourth frequency scanning curve based on the fourth frequency scanning signal; calculating a difference value between the fourth frequency sweep curve and the second frequency sweep curve to obtain a vibration parameter representing the influence of vibration on the performance of the loudspeaker; calculating a difference value between the third frequency sweep curve and the first frequency sweep curve to obtain a total vibration parameter representing the influence of vibration on the performance of the loudspeaker and the performance of the microphone array; calculating a difference value between the total vibration parameter and the vibration parameter of the loudspeaker to obtain a vibration parameter representing the influence of vibration on the performance of the microphone array;

2. The method of claim 1, wherein the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the performing parameter calculation based on the audio signal to obtain the device performance parameter includes:

3. The method of claim 2, wherein determining, for a single frequency signal at each voltage level, a sensitivity at the voltage level of a pickup channel conveying the audio signal from the single frequency signal comprises:

4. The method of claim 3, wherein the test tone is a sequence of single-frequency sinusoidal signals, the voltage is sequentially decreased from an initial voltage level to a preset voltage level with a preset tolerance, and every two adjacent groups of signals are separated by a mute signal; the ratio of the voltage of the single-frequency signal to the target sound pressure is calculated to obtain the sensitivity of the pickup channel under the voltage level, and the method comprises the following steps:

5. The method of claim 1, wherein the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the performing parameter calculation based on the audio signal to obtain the device performance parameter includes:

6. The method of claim 5, wherein determining the performance of the speech device according to the device performance parameters and outputting a performance test result comprises:

7. The method of claim 1, further comprising:

acquiring equipment performance parameters of a plurality of voice equipment;

8. A test system, comprising: the device comprises a test preparation module, a test module and a test result output module;

the testing module is used for executing the testing method of any one of the preceding claims 1-7.

9. An apparatus for testing a speech device, the speech device comprising a speaker and an array of microphones, the apparatus comprising:

a receiving module configured to receive an audio signal, the audio signal comprising: the method comprises the steps that a first audio signal collected by a microphone array when a loudspeaker plays a test audio according to a preset volume, a second audio signal collected by a standard microphone when the loudspeaker plays the test audio according to the preset volume, a third audio signal collected by the microphone array when a high-fidelity loudspeaker plays the test audio according to the same sound pressure level, and a fourth audio signal collected by the standard microphone when the high-fidelity loudspeaker plays the test audio according to the same sound pressure level;

the calculation module comprises:

the first calculation submodule is configured to calculate a difference value between the fourth frequency sweep curve and the second frequency sweep curve to obtain a vibration parameter representing the influence of vibration on the performance of the loudspeaker;

a second calculation submodule configured to calculate a difference between the third frequency sweep curve and the first frequency sweep curve to obtain a total vibration parameter representing an influence of vibrating the speaker performance and the microphone array performance;

a third calculation submodule configured to calculate a difference between the total vibration parameter and the vibration parameter of the speaker, so as to obtain a vibration parameter representing an influence of vibration on the performance of the microphone array;

10. The apparatus of claim 9, wherein the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the calculation module comprises:

11. The apparatus of claim 10, wherein the first determining submodule comprises:

12. The apparatus of claim 11, wherein the test tone is a sequence of single-frequency sinusoidal signals, the voltage is sequentially decreased from an initial voltage level to a preset voltage level with a preset tolerance, and every two adjacent groups of signals are separated by a mute signal;

13. The apparatus of claim 9, wherein the audio signal comprises: playing an audio signal collected by each microphone in the microphone array in the test audio process by the loudspeaker according to a preset volume; the calculation module comprises:

14. The apparatus of claim 13, wherein:

the determining module is configured to determine whether an echo path of the voice device meets a preset condition according to a sweep frequency curve of the voice device, and output a determination result.

15. The apparatus of claim 9, further comprising:

16. A non-transitory computer readable storage medium, on which a computer program is stored, which, when executed by a processor, performs the steps of the method of any one of claims 1 to 7.

17. A testing device for a speech device, the testing device being configured to test a speech device comprising a speaker and an array of microphones, the testing device comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to: