CN113094223A - Method for detecting voice recognition module audio channel hardware function - Google Patents
Method for detecting voice recognition module audio channel hardware function Download PDFInfo
- Publication number
- CN113094223A CN113094223A CN202110397861.8A CN202110397861A CN113094223A CN 113094223 A CN113094223 A CN 113094223A CN 202110397861 A CN202110397861 A CN 202110397861A CN 113094223 A CN113094223 A CN 113094223A
- Authority
- CN
- China
- Prior art keywords
- recognition module
- voice recognition
- test
- test program
- analog audio
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/26—Functional testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2205—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested
- G06F11/2221—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested to test input/output devices or peripheral units
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Monitoring And Testing Of Exchanges (AREA)
Abstract
A method for detecting hardware functions of an audio channel of a speech recognition module comprises the following steps: s1, writing a test program which can directly run in a built-in SRAM of a voice recognition module; s2, starting a test, wherein the test equipment firstly transmits a test program to an SRAM (static random access memory) in the voice recognition module to be tested and enables the test program to start running in the SRAM; s3, running a test program, sequentially outputting a series of analog audio electrical signals with different fixed frequencies covering low frequency, medium frequency and high frequency by a driving module, sending the analog audio electrical signals to test equipment, and inputting the analog audio electrical signals to an analog audio input port of a voice recognition module by the test equipment; and S4, judging whether the converted result digital signal is consistent with an expected result. The invention carries out loop detection through the audio output and input channels of the voice recognition module, can realize the detection function without adding a corresponding audio acquisition testing device to the testing equipment, and reduces the testing cost.
Description
Technical Field
The invention belongs to the technical field of voice recognition, and particularly relates to a method for detecting hardware functions of an audio channel of a voice recognition module.
Background
Speech recognition control technology is currently being used in a wide variety of products. Especially, with the appearance of the intelligent voice chip of the terminal, voice recognition can be realized through the voice recognition module loaded with the intelligent voice chip, voice recognition and control can be realized without being connected to a cloud server, the dependence of equipment control on a network can be removed, the real-time performance and the recognition effect of control can be guaranteed to the maximum extent, and the voice recognition control method becomes a main scheme of voice recognition control.
The realization of local voice recognition depends on a voice recognition module, the voice recognition module generally collects the sound in the environment through an audio acquisition device such as a microphone and converts the sound into an electric signal, then the unnecessary interference signal is filtered and input to an intelligent voice chip through a filter circuit consisting of passive components such as a resistor, a capacitor and an inductor on the module, the intelligent voice chip is subjected to recognition processing to control equipment, and the user is generally informed of the completion of the control through broadcasting voice feedback; in the voice feedback process, the feedback is usually that the intelligent voice chip outputs the broadcast voice corresponding to the result after the recognition processing to a loudspeaker for broadcasting after passing through a filtering or power amplifier gain circuit composed of passive components such as a resistor, a capacitor and an inductor.
In the actual production of speech recognition module, because passive components such as present resistance, electric capacity, inductance have adopted the dress formula encapsulation in a large number, the parameter value of these passive components is difficult to distinguish from the device surface to the dress formula encapsulation, and to the paster welding of passive components of every type or chip module, need artifical input material kind earlier or program or change the material to the chip mounter, artificial participation often can cause the characteristic value of making a mistake certain component, can cause the influence to the speech recognition filter effect of module after pasting like this, the gain effect of broadcast, finally can influence the use experience of product and feel. Meanwhile, if a certain passive component is welded badly during patch production, the characteristics of a filter circuit or a gain circuit are also affected, so that the recognition effect or the broadcasting effect is poor or even can not be realized, and in order to ensure the normal realization of the function of the voice recognition module, a test method capable of detecting the hardware state of the voice recognition module in batches needs to be developed.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention discloses a method for detecting the hardware function of an audio channel of a voice recognition module.
The invention discloses a method for detecting the hardware function of an audio channel of a voice recognition module, which comprises the following steps:
s1, writing a test program which can directly run in the built-in SRAM of the voice recognition module;
the test program prestores a series of analog audio electrical signals with different frequencies and corresponding expected results after identification;
the test program also comprises a driving module for sending out the analog audio electrical signals in sequence and a comparison module for comparing an expected result with an identification result;
s2, starting a test, wherein the test equipment firstly transmits a test program to an SRAM in the voice recognition module to be tested and enables the test program to start running in the SRAM;
s3, the test program runs, the drive module outputs a series of analog audio electrical signals covering different fixed frequencies of low frequency, medium frequency and high frequency in sequence, the analog audio electrical signals are sent to the test equipment, and then the test equipment inputs the analog audio electrical signals to the analog audio input port of the voice recognition module;
and S4, converting the received sequentially output fixed-frequency analog audio electrical signals into result digital signals through an ADC (analog to digital converter) by an analog audio input port of the voice recognition module, and sequentially comparing the result digital signals by a test program comparison module operated in the voice recognition chip to judge whether the converted result digital signals are consistent with an expected result.
Preferably, in S3, the test program runs in the SRAM by means of a test instruction executed by the central processing unit of the speech recognition module.
Preferably, the comparison module in S4 compares the result digital signal with the expected result to compare the frequency and amplitude of the result digital signal with the expected result.
The method for detecting the hardware function of the voice recognition module audio channel has the following advantages that:
the loop detection is carried out through the audio output and input channels of the voice recognition module, so that the detection function can be realized when a corresponding audio acquisition testing device is not added to the testing equipment, and the testing cost is reduced;
by setting different fixed frequency signals, the response of various gain and filter circuits on the module at different frequencies such as low frequency, medium frequency, high frequency and the like can be tested, if the characteristics of the circuit are changed due to the fact that the gain and filter circuits weld wrong passive components, the obtained result is different from an expected value set in a test program, corresponding errors can be detected, and the problem that the selection of the characteristic value of the passive components on the module is wrong or poor in welding cannot be directly checked by the appearance is solved;
the invention has simple test procedure and less time consumption, and reduces the time cost for testing the passive component.
Drawings
FIG. 1 is a flow chart illustrating an embodiment of detecting hardware functionality of an audio channel of a speech recognition module according to the present invention;
fig. 2 is a schematic diagram of an embodiment of a method for detecting hardware functions of an audio channel of a speech recognition module according to the present invention.
Detailed Description
The following provides a more detailed description of the present invention.
The method for detecting the hardware function of the audio channel of the voice recognition module, as shown in fig. 1, comprises the following steps:
s1, writing a test program which can directly run in the built-in SRAM of the voice recognition module;
the test program prestores a series of analog audio electrical signals with different frequencies and corresponding expected results after identification;
the test program also comprises a driving module for sending out the analog audio electrical signals in sequence and a comparison module for comparing an expected result with an identification result;
s2, starting a test, wherein the test equipment firstly transmits a test program to an SRAM in the voice recognition module to be tested and enables the test program to start running in the SRAM;
s3, the test program runs, a series of analog audio electrical signals covering different fixed frequencies of low frequency, medium frequency and high frequency are output in sequence and sent to the test equipment, and then the test equipment inputs the analog audio electrical signals to the analog audio input port of the voice recognition module;
in the embodiment shown in fig. 1, the digital audio signal sent by the test program is converted into an analog audio electrical signal by the DAC, and the analog audio electrical signal is sent to the test equipment and then is transmitted back to the ADC.
As shown in fig. 1, the passive components are generally distributed before the input end of the ADC and after the output end of the DAC, the test program runs in the SRAM by means of the test instructions run by the CPU, the CPU controls the ADC and the DAC to send the data embedded in the control program in the SDRAM, and the CPU acquires the data output by the ADC and the DAC and stores the data in the SRAM, and compares the data with the expected result of the test program stored in the SRAM.
The DAC and the ADC can be independent devices on the voice recognition module to be tested, and can also be modules integrated by a voice recognition chip.
S4, converting received sequentially output fixed-frequency analog audio electrical signals into digital signals through an analog-to-digital converter (ADC) by an analog audio input port of the voice recognition module, sequentially comparing the digital signals by a test program running in a voice recognition chip, and observing whether the digital signals are consistent with an expected result;
if the comparison result of each analog audio signal is consistent, the characteristic value and the connectivity of the passive component related to the audio channel on the module can be judged to be correct; and if the inconsistency exists, the test equipment is informed of error reporting for overhauling.
And the expected result is obtained by the test program in the test program, and then is compared with the expected result pre-stored in the test program in sequence after the test program obtains the converted result digital signal.
For example, a test program outputs a 1KHz sine wave signal, the ADC samples the signal at a sampling frequency of 16KHz, as shown in fig. 1, a data packet of the sine wave signal is obtained after the sampling is performed by a passive component and a test device, and the data packet is used as a result digital signal of the 1KHz sine wave signal, and is compared with an expected result which is obtained after the 1KHz sine wave signal prestored in the test program is sampled by 16KHz, and if the result is different, it indicates that there is an element performance or a connection relation error on the voice recognition module to be tested.
The specific comparison method may be to compare the frequency and the amplitude, if the frequency of the digital signal obtained after the sine wave is sampled is found to be the same, and the amplitude becomes smaller, it may be a problem of gain, if the frequency is found to be inconsistent, it may be a problem in the filtering part, and if the signal is found to be even unable to be sampled, it indicates that the device in the whole path may be faulty soldered and the path is disconnected.
According to the invention, loop detection is carried out through the audio output and input channels of the voice recognition module, so that the voice detection function can be realized under the condition that the test equipment has no audio acquisition test device, and the test cost is reduced;
secondly, by setting different fixed frequency signals, the invention can test the response of various gain and filter circuits on the module at different frequencies such as low frequency, intermediate frequency, high frequency and the like, if the characteristics of the circuit are changed due to the fact that the gain and filter circuits weld wrong passive components, the obtained result is different from the expected value set in the test program, the corresponding error can be detected, and the problem that the selection error or poor welding of the characteristic value of the passive components on the module cannot be directly checked by the appearance is solved;
the test procedure is relatively simple, consumes less time, and can also reduce the time cost required for testing the problem.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
Production testing was performed for a CI1102 speech recognition module, where CI1102 was a local speech recognition chip with built-in SRAM developed by the applicant at his own discretion.
The method comprises the steps of preparing a test program, sequentially sending out audio sine wave signals with fixed frequencies of 200Hz, 1KHz, 5KHz, 10KHz, 15KHz and 20KHz during test of the program, wherein the test equipment is a special test fixture for the CI1102, and the fixture can be communicated with the CI1102 voice recognition module. During testing, the CI1102 voice recognition module to be tested is placed on the testing jig and pressed tightly, after the testing function of the testing jig is started, the testing jig transmits the pre-loaded testing program to the SRAM in the CI1102 voice recognition chip in the voice recognition module to be tested, and the program starts to run in the SRAM of the chip.
When the program runs, the CI1102 voice recognition chip is activated to send out a 200Hz analog audio sine wave signal, and the signal is output to the test fixture through the audio output filtering and gain circuit on the module; the test fixture inputs the signal to the audio input port of the voice recognition module, the signal is sent to the CI1102 voice recognition chip through a filter circuit connected with the audio input port on the module, the CI1102 voice recognition chip converts the signal into a digital signal, the digital signal is compared with a value obtained when the peripheral components and parts are normal when 200Hz expected to pass through the module in the test program, if the digital signal is normal, signals of 1KHz, 5KHz, 10KHz, 15KHz and 20KHz are sequentially sent according to the steps and compared until all the signals are compared, and the test program informs the test fixture that the module to be tested is normal; if one of the comparison fails, the test program informs the test fixture that the module to be tested is abnormal, and the user can analyze and repair the module according to the problem.
The method for detecting the hardware function of the voice recognition module audio channel has the following advantages that:
the loop detection is carried out through the audio output and input channels of the voice recognition module, so that the detection function can be realized when a corresponding audio acquisition testing device is not added to the testing equipment, and the testing cost is reduced;
by setting different fixed frequency signals, the response of various gain and filter circuits on the module at different frequencies such as low frequency, medium frequency, high frequency and the like can be tested, if the characteristics of the circuit are changed due to the fact that the gain and filter circuits weld wrong passive components, the obtained result is different from an expected value set in a test program, corresponding errors can be detected, and the problem that the selection of the characteristic value of the passive components on the module is wrong or poor in welding cannot be directly checked by the appearance is solved;
the invention has simple test procedure and less time consumption, and reduces the time cost for testing the passive component.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is directed to preferred embodiments of the present invention, wherein the preferred embodiments are not obviously contradictory or subject to any particular embodiment, and any combination of the preferred embodiments may be combined in any overlapping manner, and the specific parameters in the embodiments and examples are only for the purpose of clearly illustrating the inventor's invention verification process and are not intended to limit the scope of the invention, which is defined by the claims and the equivalent structural changes made by the description and drawings of the present invention are also intended to be included in the scope of the present invention.
Claims (3)
1. A method for detecting hardware functions of an audio channel of a speech recognition module, comprising the steps of:
s1, writing a test program which can directly run in the built-in SRAM of the voice recognition module;
the test program prestores a series of analog audio electrical signals with different frequencies and corresponding expected results after identification;
the test program also comprises a driving module for sending out the analog audio electrical signals in sequence and a comparison module for comparing an expected result with an identification result;
s2, starting a test, wherein the test equipment firstly transmits a test program to an SRAM in the voice recognition module to be tested and enables the test program to start running in the SRAM;
s3, the test program runs, the drive module outputs a series of analog audio electrical signals covering different fixed frequencies of low frequency, medium frequency and high frequency in sequence, the analog audio electrical signals are sent to the test equipment, and then the test equipment inputs the analog audio electrical signals to the analog audio input port of the voice recognition module;
and S4, converting the received sequentially output fixed-frequency analog audio electrical signals into result digital signals through an ADC (analog to digital converter) by an analog audio input port of the voice recognition module, and sequentially comparing the result digital signals by a test program comparison module operated in the voice recognition chip to judge whether the converted result digital signals are consistent with an expected result.
2. The method for detecting hardware functionality of an audio channel of a speech recognition module as claimed in claim 1, wherein in S3, the test program is executed in the SRAM by means of a test instruction executed by the central processing unit of the speech recognition module.
3. The method for detecting hardware functionality of an audio channel of a speech recognition module according to claim 1, wherein the comparison of the resulting digital signal and the expected result by the comparison module in S4 is a comparison of frequency and amplitude of the two.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397861.8A CN113094223B (en) | 2021-04-14 | 2021-04-14 | Method for detecting hardware function of voice recognition module audio channel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397861.8A CN113094223B (en) | 2021-04-14 | 2021-04-14 | Method for detecting hardware function of voice recognition module audio channel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113094223A true CN113094223A (en) | 2021-07-09 |
| CN113094223B CN113094223B (en) | 2023-05-02 |
Family
ID=76677106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110397861.8A Active CN113094223B (en) | 2021-04-14 | 2021-04-14 | Method for detecting hardware function of voice recognition module audio channel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113094223B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116705069A (en) * | 2023-07-31 | 2023-09-05 | 北京芯驰半导体科技有限公司 | Audio detection method, chip, component on traffic equipment and traffic equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197836A1 (en) * | 2004-01-08 | 2005-09-08 | Jordan Cohen | Automated testing of voice recognition software |
| CN101154175A (en) * | 2006-09-25 | 2008-04-02 | 佛山市顺德区顺达电脑厂有限公司 | Audio testing method and system combining with voice recognition |
| CN101575011A (en) * | 2009-05-05 | 2009-11-11 | 中国民航大学 | Automatic test equipment of boeing series aircraft audio management unit and method thereof |
| US20150162018A1 (en) * | 2013-12-09 | 2015-06-11 | Honeywell International Inc. | Voice based diagnostic systems and methods |
| CN104931823A (en) * | 2015-06-08 | 2015-09-23 | 小米科技有限责任公司 | Electronic device test method and device |
| CN108872834A (en) * | 2018-06-29 | 2018-11-23 | 上海与德通讯技术有限公司 | The test macro and method of voice recognition chip, voice recognition chip |
| CN109979434A (en) * | 2019-04-30 | 2019-07-05 | 成都启英泰伦科技有限公司 | The test method of local voice mould set product acoustical behavior |
| CN110446150A (en) * | 2019-08-07 | 2019-11-12 | 厦门市思芯微科技有限公司 | A kind of intelligence Wi-Fi speaker automated production test macro and method |
| CN111354335A (en) * | 2018-12-24 | 2020-06-30 | 深圳市优必选科技有限公司 | Voice recognition testing method and device, storage medium and terminal equipment |
| CN112365883A (en) * | 2020-10-29 | 2021-02-12 | 安徽江淮汽车集团股份有限公司 | Cabin system voice recognition test method, device, equipment and storage medium |
-
2021
- 2021-04-14 CN CN202110397861.8A patent/CN113094223B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050197836A1 (en) * | 2004-01-08 | 2005-09-08 | Jordan Cohen | Automated testing of voice recognition software |
| CN101154175A (en) * | 2006-09-25 | 2008-04-02 | 佛山市顺德区顺达电脑厂有限公司 | Audio testing method and system combining with voice recognition |
| CN101575011A (en) * | 2009-05-05 | 2009-11-11 | 中国民航大学 | Automatic test equipment of boeing series aircraft audio management unit and method thereof |
| US20150162018A1 (en) * | 2013-12-09 | 2015-06-11 | Honeywell International Inc. | Voice based diagnostic systems and methods |
| CN104931823A (en) * | 2015-06-08 | 2015-09-23 | 小米科技有限责任公司 | Electronic device test method and device |
| CN108872834A (en) * | 2018-06-29 | 2018-11-23 | 上海与德通讯技术有限公司 | The test macro and method of voice recognition chip, voice recognition chip |
| CN111354335A (en) * | 2018-12-24 | 2020-06-30 | 深圳市优必选科技有限公司 | Voice recognition testing method and device, storage medium and terminal equipment |
| CN109979434A (en) * | 2019-04-30 | 2019-07-05 | 成都启英泰伦科技有限公司 | The test method of local voice mould set product acoustical behavior |
| CN110446150A (en) * | 2019-08-07 | 2019-11-12 | 厦门市思芯微科技有限公司 | A kind of intelligence Wi-Fi speaker automated production test macro and method |
| CN112365883A (en) * | 2020-10-29 | 2021-02-12 | 安徽江淮汽车集团股份有限公司 | Cabin system voice recognition test method, device, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| SHALINI A等: "Voice Recognition Robot Control Using Android Device", 《JOURNAL OF COMPUTATIONAL AND THEORETICAL NANOSCIENCE》 * |
| 万声国: "基于语音识别的智能WIFI音箱", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116705069A (en) * | 2023-07-31 | 2023-09-05 | 北京芯驰半导体科技有限公司 | Audio detection method, chip, component on traffic equipment and traffic equipment |
| CN116705069B (en) * | 2023-07-31 | 2023-11-10 | 北京芯驰半导体科技有限公司 | Audio detection method, chip, component on traffic equipment and traffic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113094223B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8879744B2 (en) | Audio testing system and audio testing method for device under test | |
| US9439010B2 (en) | System for tuning audio processing features and method thereof | |
| CN108511000B (en) | Method and system for testing identification rate of awakening words of intelligent sound box | |
| CN104581605B (en) | The system and method calibrated automatically for energy converter | |
| CN108124042B (en) | Device fault detection method and device and mobile terminal | |
| CN108430026B (en) | Audio equipment fault detection method and karaoke equipment | |
| CN110072166A (en) | A kind of hardware adjustment method of digital microphone | |
| CN108347686A (en) | Audio testing method, device, smart machine and storage medium | |
| CN101917735A (en) | Mobile terminal audio calibrating method and automatic testing system | |
| CN113094223A (en) | Method for detecting voice recognition module audio channel hardware function | |
| CN210294434U (en) | Multifunctional automatic test system of vehicle-mounted multimedia system | |
| CN103581819A (en) | Microphone detection method | |
| CN108064347A (en) | The audio pin test method and test device of communication module | |
| US6370252B1 (en) | Circuit arrangement for testing the connection of a sound reproducing device to a sound signal source | |
| CN107682787A (en) | Speakers fit method and apparatus for vehicle audio | |
| CN111787479B (en) | Method and system for correcting listening sensation of TWS earphone | |
| US6501277B1 (en) | Automotive harness and audio system analyzer | |
| CN207995410U (en) | A kind of bluetooth conversation performance detecting system | |
| KR101473616B1 (en) | Feedback Apparatus of Audio Signal For Handphone and Transfer Method of Sound Source using the Apparatus | |
| US20090220097A1 (en) | Sound testing device for mobile phone and method for using the same | |
| CN110113701B (en) | Hardware debugging method for analog microphone | |
| CN114676043A (en) | Testing method and device of intelligent voice module, storage medium and electronic device | |
| CN210609714U (en) | Automatic testing component for microphone recording port | |
| CN113518295A (en) | Hearing aid chip test system and test method thereof | |
| CN112399323A (en) | Sound box audio automatic test comparison device, system and test method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |