CN108597494A - Tone testing method and device - Google Patents
Tone testing method and device Download PDFInfo
- Publication number
- CN108597494A CN108597494A CN201810188414.XA CN201810188414A CN108597494A CN 108597494 A CN108597494 A CN 108597494A CN 201810188414 A CN201810188414 A CN 201810188414A CN 108597494 A CN108597494 A CN 108597494A
- Authority
- CN
- China
- Prior art keywords
- measured
- voice plate
- wake
- test
- word
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/01—Assessment or evaluation of speech recognition systems
Abstract
The invention discloses a kind of tone testing method and devices.Wherein, this method includes：Obtain audio test file；According to the state of voice plate to be measured test assignment is executed by playing the audio test file to the voice plate to be measured；Generate the test result for being used to indicate the voice plate performance to be measured.The present invention is solved takes longer technical problem since existing tone testing method needs multiple testers to carry out test caused by prolonged thought instructs.
Description
Technical field
The present invention relates to tone testing fields, in particular to a kind of tone testing method and device.
Background technology
For the needs of some phonological components exploitation, the reliability of its voice is verified after completing estimated function, needed
It is tested for the property.Current speech performance test method needs 10 or 10 or more testers to carry out for a long time
Thought instruction, the wake-up rate and discrimination of voice plate are tested for the property, time-consuming, is easy to keep tester tired, cause
Test result has error.
For above-mentioned problem, currently no effective solution has been proposed.
Invention content
An embodiment of the present invention provides a kind of tone testing method and devices, at least to solve due to existing tone testing
Test caused by method needs multiple testers to carry out prolonged thought instruction takes longer technical problem.
One side according to the ... of the embodiment of the present invention provides a kind of tone testing method, including：Obtain audio-frequency test text
Part；It executes test by playing the audio test file to the voice plate to be measured according to the state of voice plate to be measured and appoints
Business；Generate the test result for being used to indicate the voice plate performance to be measured.
Optionally, the audio test file includes the first audio file for waking up test, wherein according to language to be measured
The state of soundboard, by playing the audio test file to the voice plate to be measured, executing test assignment includes：It is waited for described
The state for surveying voice plate is in the case of waiting for wake-up states, by play first audio file to the voice plate to be measured into
Row wakes up test, obtains the wake-up rate of the voice plate to be measured.
Optionally, described before carrying out wake-up test to the voice plate to be measured by broadcasting first audio file
Method further includes：Obtain the state of the voice plate to be measured；Detect the state of the voice plate to be measured；If the voice plate to be measured
State be it is described wait for wake-up states, triggering, which executes the mistake and plays first audio file, carries out the voice plate to be measured
Wake up test；If the state of the voice plate to be measured is wake-up states, by playing third audio text to the voice plate to be measured
Part controls the voice plate to be measured and exits the wake-up states；If the state of the voice plate to be measured is information source broadcast state, lead to
It crosses and the 4th audio file is played to the voice plate to be measured, the control voice plate to be measured stops playing information source, and is waited for described
It surveys voice plate to stop after playing information source, by playing the third audio file to the voice plate to be measured, be waited for described in control
It surveys voice plate and exits the wake-up states.
Optionally, wake-up test is carried out to the voice plate to be measured by playing first audio file, obtained described
The wake-up rate of voice plate to be measured includes：The voice plate to be measured is played in first audio file by Hi-Fi sound-box
Wake up word；Obtain the first word that the voice plate to be measured returns；Judge whether first word is the wake-up word；If institute
It is the wake-up word to state the first word, and determination wakes up successfully；According to successful number is waken up, the wake-up rate is obtained.
Optionally, the method further includes：If first word is not the wake-up word, the wake-up is played again
Word；Obtain the second word that the voice plate to be measured returns；If second word is the wake-up word, determination wakes up successfully.
Optionally, the method further includes：If second word is not the wake-up word, repeat playing for forcing to call out
The fifth audio of the awake voice plate to be measured；If successfully waking up the voice plate to be measured in preset times, continue described
Wake up test；If not waking up the voice plate to be measured more than the preset times, terminate the wake-up test.
Optionally, the audio test file includes the second audio file tested for identification and second audio text
The corresponding mark file of part, wherein according to the state of voice plate to be measured, surveyed by playing the audio to the voice plate to be measured
File is tried, executing test assignment includes：In the case where the state of the voice plate to be measured is wake-up states, according to described second
File is marked described in audio file, test is identified to the voice plate to be measured, obtain the discrimination of the voice plate to be measured.
Optionally, test is identified to the voice plate to be measured according to mark file described in second audio file,
The discrimination for obtaining the voice plate to be measured includes：Second audio is played to the voice plate to be measured by Hi-Fi sound-box
Instruction word in file；Obtain the word content that the voice plate to be measured returns；Judge the word content and mark text
Whether the information indicated in part is identical；If the word content is identical as the information that the mark document is shown, identification is determined
Success；According to successful number is identified, the discrimination is obtained.
Optionally, the method further includes：If the information that the word content shows with the mark document differs,
Described instruction word is played again；If the letter that the word content that the voice plate to be measured again returns to shows with the mark document
Manner of breathing is same, and determination identifies successfully；If the word content that the voice plate to be measured again returns to shows with the mark document
Information differs, and determines recognition failures.
Optionally, the mark file includes at least one of：Audio file name, voice content, format.
Another aspect according to the ... of the embodiment of the present invention additionally provides a kind of tone testing device, including：Acquiring unit is used
In acquisition audio test file；Test cell, for the state according to voice plate to be measured, by being played to the voice plate to be measured
The audio test file executes test assignment；Generation unit, for generating the survey for being used to indicate the voice plate performance to be measured
Test result.
Another aspect according to the ... of the embodiment of the present invention, additionally provides a kind of storage medium, and the storage medium includes storage
Program, wherein described program executes above-mentioned tone testing method.
Another aspect according to the ... of the embodiment of the present invention additionally provides a kind of terminal, including：Memory, for depositing
Store up program；Processor, for running described program, wherein described program executes above-mentioned tone testing method when running.
In embodiments of the present invention, using acquisition audio test file；According to the state of voice plate to be measured, by described
Voice plate to be measured plays the audio test file, executes test assignment；Generation is used to indicate the voice plate performance to be measured
The mode of test result has reached voice by carrying out automatic speech performance test, statistics according to the state of voice plate to be measured
Can test automation, Unmanned operation purpose, to realize reduce tester, accelerate test speed technology effect
Fruit, and then solve since existing tone testing method needs multiple testers surveyed caused by prolonged thought instructs
Examination takes longer technical problem.
Description of the drawings
Attached drawing described herein is used to provide further understanding of the present invention, and is constituted part of this application, this hair
Bright illustrative embodiments and their description are not constituted improper limitations of the present invention for explaining the present invention.In the accompanying drawings：
Fig. 1 is a kind of flow diagram of optional tone testing method according to the ... of the embodiment of the present invention；
Fig. 2 is the flow diagram of another optional tone testing method according to the ... of the embodiment of the present invention；
Fig. 3 is a kind of optional software interface schematic diagram according to the ... of the embodiment of the present invention；
Fig. 4 is the flow diagram of another optional tone testing method according to the ... of the embodiment of the present invention；
Fig. 5 is the flow diagram of another optional tone testing method according to the ... of the embodiment of the present invention；
Fig. 6 is the flow diagram of another optional tone testing method according to the ... of the embodiment of the present invention；
Fig. 7 is a kind of structural schematic diagram of optional tone testing device according to the ... of the embodiment of the present invention.
Specific implementation mode
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The every other embodiment that member is obtained without making creative work should all belong to the model that the present invention protects
It encloses.
It should be noted that term " first " in description and claims of this specification and above-mentioned attached drawing, "
Two " etc. be for distinguishing similar object, without being used to describe specific sequence or precedence.It should be appreciated that using in this way
Data can be interchanged in the appropriate case, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
It includes to be not necessarily limited to for example, containing the process of series of steps or unit, method, system, product or equipment to cover non-exclusive
Those of clearly list step or unit, but may include not listing clearly or for these processes, method, product
Or the other steps or unit that equipment is intrinsic.
Embodiment 1
According to embodiments of the present invention, a kind of embodiment of the method for tone testing method is provided, it should be noted that attached
The step of flow of figure illustrates can execute in the computer system of such as a group of computer-executable instructions, though also,
So logical order is shown in flow charts, but in some cases, it can be with different from shown by sequence execution herein
Or the step of description.
Fig. 1 is tone testing method according to the ... of the embodiment of the present invention, as shown in Figure 1, this method comprises the following steps：
Step S102 obtains audio test file.
The tone testing method of the present embodiment, test need to rely on audio test file, audio test file institute it is right
The mark file answered, audio test file include to force to wake up audio file, force to stop playing audio file, positive closing
Speech audio file and test identification and the required sample audio file of wake-up；It is then speech recognition to be tested to mark file
Information (filename containing testing audio, the broadcasting content of testing audio file, testing audio text corresponding to the audio file of project
Format information corresponding to part), it so that computer is obtained voice plate and pass through the information and mark text that serial ports is returned
Information contained by part is compared.
Step S104 executes survey according to the state of voice plate to be measured by playing audio test file to voice plate to be measured
Trial is engaged in.
As a kind of optional realization method, audio test file includes for waking up the first audio file of test (i.e.
Wake up word audio file), wherein it is held according to the state of voice plate to be measured by playing audio test file to voice plate to be measured
Row test assignment includes：In the case where the state of voice plate to be measured is to wait for wake-up states, by playing the first audio file pair
Voice plate to be measured carries out wake-up test, obtains the wake-up rate of voice plate to be measured.
Optionally, before carrying out wake-up test to voice plate to be measured by the first audio file of broadcasting, method further includes：It obtains
Take the state of voice plate to be measured；Detect the state of voice plate to be measured；If the state of voice plate to be measured is to wait for that wake-up states, triggering are held
The first audio file of broadcasting of going carries out wake-up test to voice plate to be measured；If the state of voice plate to be measured is wake-up states, lead to
It crosses and third audio file is played to voice plate to be measured, control voice plate to be measured and exit wake-up states；If the state of voice plate to be measured
For information source broadcast state, by playing the 4th audio file to voice plate to be measured, controls voice plate to be measured and stop playing information source, and
After voice plate to be measured stops playing information source, by playing third audio file to voice plate to be measured, voice plate to be measured is controlled
Exit wake-up states.
The tone testing method of the present embodiment, before testing a phonetic order, voice plate necessarily is in shape to be waken up
State meets the principle that actual user uses just now.Automatic test software, needs whether detection voice plate is in shape to be waken up first
State allows computer to play positive closing voice to voice plate by Hi-Fi sound-box if the state of voice plate is wake-up states
Audio file (i.e. above-mentioned third audio file), to allow it to exit wake-up states；If voice plate is in information source broadcast state,
It allows computer to play voice plate by Hi-Fi sound-box to force to stop to play audio file (the 4th i.e. above-mentioned audio file),
It is to be obtained to stop playing the state of information source to voice plate and then play positive closing speech audio file (i.e. above-mentioned third
Audio file), to allow it to exit wake-up states.
Optionally, wake-up test is carried out to voice plate to be measured by playing the first audio file, obtains voice plate to be measured
Wake-up rate includes：The wake-up word in the first audio file is played to voice plate to be measured by Hi-Fi sound-box；Obtain voice to be measured
The first word that plate returns；Judge whether the first word is to wake up word；If the first word is to wake up word, determination wakes up successfully；Root
According to successful number is waken up, wake-up rate is obtained.
Optionally, the tone testing method of the present embodiment further includes：If the first word is not to wake up word, plays wake up again
Word；Obtain the second word that voice plate to be measured returns；If the second word is to wake up word, determination wakes up successfully.
Optionally, the tone testing method of the present embodiment further includes：If the second word is not to wake up word, repeats playing and be used for
Force the fifth audio of wake-up voice plate to be measured；If successfully waking up voice plate to be measured in preset times, continues to wake up and survey
Examination；If not waking up voice plate to be measured more than preset times, terminate to wake up test.
The wake-up rate of the present embodiment tested speech plate need to be in voice plate and wait for that wake-up states carry out.Ensure that voice plate exists
It waits for wake-up states, allows computer by Hi-Fi sound-box to play different wake-up word audio texts to voice plate at this time (i.e. above-mentioned
First audio file) part judges to return after waiting voices plate to return to the word corresponding to the audio signal that gets by serial ports
Word content whether be wake up word, if so, wake up number of success add 1；If not word is waken up, or wake-up word is not right, then
Current audio file is repeated playing again 1 time；If success, wakes up number of success and adds 1, if wake-up is unsuccessful, wake up failure
Number adds 1；Voice plate of not waking up all is called out when playing twice, then repeats playing pressure and wakes up audio file (i.e. above-mentioned fifth audio
File), number of repetition 10 times then continues task as long as 10 interior energies wake up voice plate；When more than 10 times strong wake up, then appoint
Business terminates, software prompt " preset test environment is wrong, asks inspection environment ".
As another optional realization method, audio test file include the second audio file tested for identification and
The corresponding mark file of second audio file, wherein according to the state of voice plate to be measured, by playing audio to voice plate to be measured
Test file, executing test assignment includes：In the case where the state of voice plate to be measured is wake-up states, according to the second audio text
Test is identified to voice plate to be measured in part mark file, obtains the discrimination of voice plate to be measured.
Optionally, file is marked according to the second audio file and test is identified to voice plate to be measured, obtain voice to be measured
The discrimination of plate includes：The instruction word in the second audio file is played to voice plate to be measured by Hi-Fi sound-box；It obtains to be measured
The word content that voice plate returns；Judge word content and whether mark the information shown of document identical；If word content with
The information that mark document is shown is identical, and determination identifies successfully；According to successful number is identified, discrimination is obtained.
Optionally, method further includes：If the information that word content shows with mark document differs, again play instruction
Word；If the word content that voice plate to be measured again returns to is identical as the information that mark document is shown, determination identifies successfully；If waiting for
It surveys the information that the word content that voice plate again returns to shows with mark document to differ, determines recognition failures.
The present embodiment tests discrimination, must first wake up voice plate, and after voice plate is waken up, rate survey can be identified
Examination.Discrimination also has chance twice, Hi-Fi sound-box to play different test words audio file (i.e. above-mentioned second to voice plate
Audio file), the broadcasting of word content and audio file corresponding in mark file that voice plate returns is obtained by serial ports
Whether content, JSON (JavaScript Object Notation, a kind of data interchange format of lightweight) are consistent, consistent to remember
It is inconsistent for success, it is denoted as failure；It is identified successfully when first time, then identifies that number of success adds 1, carry out next cycle, if
1st time unsuccessful, then repeats playing 1 time, fail when twice, and recognition failures number adds 1, then carries out next cycle.
Optionally, mark file includes at least one of：Audio file name, voice content, format.
Step S106 generates the test result for being used to indicate voice plate performance to be measured.
After the completion of all tasks, the wake-up number of success that software can be tested is converted to wake-up rate, identifies successfully
Number is converted to discrimination, generates final test result, then exports to Excel tables.
By the method, the progress synchronous with discrimination of wake-up rate is realized, to test the sound of different people to speech recognition
The voice signal of different people need to only be recorded into audio file while the audio file is added in file marking by the influence of performance
The information such as name, corresponding voice content and JSON.
Through the above steps, it can be reached by carrying out automatic speech performance test, statistics according to the state of voice plate to be measured
Arrived speech performance test automation, Unmanned operation purpose, to realize reduce tester, accelerate test speed
The technique effect of degree, and then solve and refer to since existing tone testing method needs multiple testers to carry out prolonged read
It is tested caused by order and takes longer technical problem.Realize speech performance test automation, Unmanned operation；Realize that standardization is surveyed
Examination, speech performance test accuracy is high, and test phenomenon can reappear；Shorten the speech performance testing time；Exclude artificial subjective factor
Interference, test overall process standardization.
In the following, as shown in Fig. 2, the tone testing method to the present embodiment illustrates：
Step A starts.
The tone testing method of the present embodiment plays audio file to simulate voice (more true to nature and people using Hi-Fi sound-box
Sound of speaking), by Hi-Fi sound-box place on the holders of 1.5 meters of height, to simulate the face position of people.Determining height
Behind the position of fidelity speaker, opens software and tested, software interface is as shown in Figure 3.
Step B, establishes task.
The software of the present embodiment can be based on running under Python2.7 environment, test the testing audio text that needs to rely on
Mark file corresponding to part, testing audio, testing audio file include to force to wake up audio file, force to stop playing
Audio file, positive closing speech audio file and test identification and the required sample audio file of wake-up；Mark file then
It is information (filename containing testing audio, the testing audio file corresponding to speech recognized terms purpose audio file to be tested
JSON information corresponding to broadcasting content, testing audio file), it so that computer is obtained voice plate and pass through serial ports institute
The information of return is compared with the information contained by mark file.Software, which carries out test, to be carried out according to the content of mark file
, how many row will loop test how many times, carried out every time as task.
Step C, is guaranteed at and waits for wake-up states.
Before testing a phonetic order, voice plate, which necessarily is in, waits for wake-up states, meets actual user's use just now
Principle.As shown in figure 4, being guaranteed at the step of waiting for wake-up states and including：
Step c1 starts.
Step c2 obtains voice board status.
Step c3, if in waiting for wake-up states.
If it is not, executing step c4；
If so, executing step c5.
Step c4 plays the audio for exiting wake-up.
Step c5 terminates.
Whether automatic test software needs detection voice plate in wake-up states are waited for, if the state of voice plate is to call out first
The state of waking up then allows computer to play positive closing speech audio file to voice plate by Hi-Fi sound-box, is called out with allowing it to exit
The state of waking up；If voice plate is in information source broadcast state, allows computer to be played to voice plate by Hi-Fi sound-box and force to stop
Audio file is played, it is to be obtained to stop playing the state of information source to voice plate and then play positive closing speech audio file,
To allow it to exit wake-up states.
Step D wakes up test.
The wake-up rate of tested speech plate need to be in voice plate and wait for that wake-up states carry out.As shown in figure 5, waking up test
Step includes：
Step d1 starts.
Step d2 is played and is waken up word audio.
Step d3 obtains voice board status.
Step d4, if be waken up.
If it is not, executing d5；
If so, executing step d11.
Step d5 is played wake up word audio again.
Step d6 obtains voice board status.
Step d7, if be waken up.
If it is not, executing step d8；
If so, executing step d11.
Step d8 wakes up the frequency of failure+1.
Step d9 plays strong wake-up word audio.
Step d10, if be waken up.
If it is not, executing step d9；
If so, executing step d12.
Step d11 wakes up number of success+1.
Step d12 terminates.
After step C, it is ensured that voice plate is waiting for wake-up states, allows computer by Hi-Fi sound-box to voice plate at this time
Play different wake-up word audio files, wait voices plate by serial ports return word corresponding to the audio signal that gets it
Afterwards, judge whether the word content returned is to wake up word, if so, waking up number of success adds 1；If not waking up word, or call out
Awake word is not right, then repeats playing current audio file again 1 time；If success, wakes up number of success and adds 1, if wake-up is unsuccessful,
It then wakes up the frequency of failure and adds 1；Voice plate of not waking up all is called out when playing twice, then repeats playing pressure and wakes up audio file, is repeated secondary
Number 10 times then continues task as long as 10 interior energies wake up voice plate；When more than 10 times strong wake up, then task terminates, and software carries
Show " preset test environment is wrong, asks inspection environment ".
Step E, identification test.
Discrimination is tested, voice plate must be first waken up, after voice plate is waken up, rate test can be identified.Such as Fig. 6
It is shown, identify that the step of testing includes：
Step e1 starts.
Step e2, play instruction word audio.
Step e3 obtains voice plate recognition result.
Step e4, if identify successfully.
If it is not, executing step e5；
If so, executing step e9.
Step e6, again play instruction word audio.
Step e7 obtains voice plate recognition result.
Step e8, if identify successfully.
If so, executing step e9；
If it is not, executing step e10.
Step e9 identifies number of success+1.
Step e10, recognition failures number+1.
Step e11 terminates.
Discrimination also has chance twice, Hi-Fi sound-box to play different test words audio files to voice plate, pass through string
Mouthful obtain the broadcasting content of word content that voice plate returns and audio file corresponding in mark file, JSON whether one
It causes, is unanimously denoted as success, it is inconsistent, it is denoted as failure；It is identified successfully when first time, then identifies that number of success adds 1, carry out next
A cycle repeats playing 1 time, fails when twice if the 1st time unsuccessful, and recognition failures number adds 1, then carries out next
Cycle.
Whether step F, task are completed.
Step G generates excel reports.
Finally all tasks are completed, the wake-up number of success that software can be tested is converted to wake-up rate, is identified as
Work(number is converted to discrimination, then exports to Excel tables.By the method, wake-up rate and the same stepping of discrimination are realized
Row, will test influence of the sound to speech recognition performance of different people, need to only record the voice signal of different people into audio text
The information such as the audio file name, corresponding voice content and JSON are added marking simultaneously in part in file.
Step H terminates.
The tone testing method of the present embodiment, using GUI (Graphical User Interface, graphic user interface)
Interface, easy to operate, easy to get started；Using computer play testing audio, by serial ports receive voice plate return data to
It is analyzed, is compared with mark file, calculate speech recognition test result, and result is exported to Excel tables, entirely
Process realizes automation；And then wake-up rate and discrimination concurrent testing are realized, the testing time is greatly shortened, test effect is improved
Rate, relative to traditional voice performance test, personnel demand quantity reduces 90%, and the testing time reduces 50%.
Embodiment 2
According to embodiments of the present invention, a kind of tone testing device is provided, as shown in fig. 7, the tone testing device includes：
Acquiring unit 702, for obtaining audio test file；Test cell 704, for the shape according to voice plate to be measured
State executes test assignment by playing the audio test file to the voice plate to be measured；Generation unit 706, for generating
It is used to indicate the test result of the voice plate performance to be measured.
Optionally, the audio test file includes the first audio file for waking up test, wherein test cell
704 state for executing following steps according to voice plate to be measured, by playing the audio-frequency test to the voice plate to be measured
File executes test assignment：In the case where the state of the voice plate to be measured is to wait for wake-up states, by playing described first
Audio file carries out wake-up test to the voice plate to be measured, obtains the wake-up rate of the voice plate to be measured.
Optionally, test cell 704 are additionally operable to obtain the state of the voice plate to be measured；Detect the voice plate to be measured
State；If the state of the voice plate to be measured waits for that wake-up states, triggering execute the mistake and play first audio described in being
File carries out wake-up test to the voice plate to be measured；If the state of the voice plate to be measured is wake-up states, by described
Voice plate to be measured plays third audio file, controls the voice plate to be measured and exits the wake-up states；If the voice to be measured
The state of plate is information source broadcast state, by playing the 4th audio file to the voice plate to be measured, controls the voice to be measured
Plate stops playing information source, and after the voice plate to be measured stops playing information source, by playing institute to the voice plate to be measured
Third audio file is stated, the voice plate to be measured is controlled and exits the wake-up states.
Optionally, test cell 704 is for executing following steps by playing first audio file to described to be measured
Voice plate carries out wake-up test, obtains the wake-up rate of the voice plate to be measured：By Hi-Fi sound-box to the voice plate to be measured
Play the wake-up word in first audio file；Obtain the first word that the voice plate to be measured returns；Judge described first
Whether word is the wake-up word；If first word is the wake-up word, determination wakes up successfully；It is successfully secondary according to waking up
Number, obtains the wake-up rate.
Optionally, it is called out described in broadcasting again if it is the wake-up word to be additionally operable to first word not test cell 704
Awake word；Obtain the second word that the voice plate to be measured returns；If second word be the wake-up word, determine wake up at
Work(.
Optionally, test cell 704 repeat playing if it is the wake-up word to be additionally operable to second word not for strong
System wakes up the fifth audio of the voice plate to be measured；If successfully waking up the voice plate to be measured in preset times, continue
The wake-up test；If not waking up the voice plate to be measured more than the preset times, terminate the wake-up test.
Optionally, the audio test file includes the second audio file tested for identification and second audio text
The corresponding mark file of part, wherein the state that test cell 704 is used to execute following steps according to voice plate to be measured, by right
The voice plate to be measured plays the audio test file, executes test assignment：It is to wake up in the state of the voice plate to be measured
In the case of state, test is identified to the voice plate to be measured according to file is marked described in second audio file, is obtained
To the discrimination of the voice plate to be measured.
Optionally, test cell 704 is for executing following steps according to mark file pair described in second audio file
Test is identified in the voice plate to be measured, obtains the discrimination of the voice plate to be measured：It is waited for described by Hi-Fi sound-box
It surveys voice plate and plays the instruction word in second audio file；Obtain the word content that the voice plate to be measured returns；Judge
Whether the word content and the information that the mark document is shown are identical；If in the word content and the mark file
The information of instruction is identical, and determination identifies successfully；According to successful number is identified, the discrimination is obtained.
Optionally, test cell 704, if being additionally operable to the word content and the information shown of mark document not phase
Together, described instruction word is played again；If the word content that the voice plate to be measured again returns to shows with the mark document
Information it is identical, determination identify successfully；If the word content that the voice plate to be measured again returns to and the mark document
The information shown differs, and determines recognition failures.
Optionally, the mark file includes at least one of：Audio file name, voice content, format.
According to embodiments of the present invention, a kind of storage medium is additionally provided, the storage medium includes the program of storage,
In, described program executes above-mentioned tone testing method.
According to embodiments of the present invention, a kind of terminal is additionally provided, including：Memory, for storing program；Processing
Device, for running described program, wherein described program executes above-mentioned tone testing method when running.
The embodiments of the present invention are for illustration only, can not represent the quality of embodiment.
In the above embodiment of the present invention, all emphasizes particularly on different fields to the description of each embodiment, do not have in some embodiment
The part of detailed description may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, for example, the unit division, Ke Yiwei
A kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, unit or module
It connects, can be electrical or other forms.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
On unit.Some or all of unit therein can be selected according to the actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of SFU software functional unit.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can be stored in a computer read/write memory medium.Based on this understanding, technical scheme of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes：USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD etc. are various can to store program code
Medium.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (13)
1. a kind of tone testing method, which is characterized in that including：
Obtain audio test file；
It executes test by playing the audio test file to the voice plate to be measured according to the state of voice plate to be measured and appoints
Business；
Generate the test result for being used to indicate the voice plate performance to be measured.
2. according to the method described in claim 1, it is characterized in that, the audio test file includes for waking up test
One audio file, wherein according to the state of voice plate to be measured, by playing the audio-frequency test text to the voice plate to be measured
Part, executing test assignment includes：
In the case where the state of the voice plate to be measured is to wait for wake-up states, by playing first audio file to described
Voice plate to be measured carries out wake-up test, obtains the wake-up rate of the voice plate to be measured.
3. according to the method described in claim 2, it is characterized in that, by playing first audio file to the language to be measured
Before soundboard carries out wake-up test, the method further includes：
Obtain the state of the voice plate to be measured；
Detect the state of the voice plate to be measured；
If the state of the voice plate to be measured waits for that wake-up states, triggering execute the mistake and play first audio file described in being
Wake-up test is carried out to the voice plate to be measured；
If the state of the voice plate to be measured is wake-up states, by playing third audio file, control to the voice plate to be measured
It makes the voice plate to be measured and exits the wake-up states；
If the state of the voice plate to be measured is information source broadcast state, by playing the 4th audio text to the voice plate to be measured
Part, the control voice plate to be measured stop playing information source, and after the voice plate to be measured stops playing information source, by institute
It states voice plate to be measured and plays the third audio file, control the voice plate to be measured and exit the wake-up states.
4. according to the method described in claim 2, it is characterized in that, by playing first audio file to the language to be measured
Soundboard carries out wake-up test, and the wake-up rate for obtaining the voice plate to be measured includes：
The wake-up word in first audio file is played to the voice plate to be measured by Hi-Fi sound-box；
Obtain the first word that the voice plate to be measured returns；
Judge whether first word is the wake-up word；
If first word is the wake-up word, determination wakes up successfully；
According to successful number is waken up, the wake-up rate is obtained.
5. according to the method described in claim 4, it is characterized in that, the method further includes：
If first word is not the wake-up word, the wake-up word is played again；
Obtain the second word that the voice plate to be measured returns；
If second word is the wake-up word, determination wakes up successfully.
6. according to the method described in claim 5, it is characterized in that, the method further includes：
If second word is not the wake-up word, the five notes of traditional Chinese music for forcing to wake up the voice plate to be measured are repeated playing
Frequently；
If successfully waking up the voice plate to be measured in preset times, continue the wake-up test；
If not waking up the voice plate to be measured more than the preset times, terminate the wake-up test.
7. according to the method described in claim 1, it is characterized in that, the audio test file includes tested for identification
Two audio files and the corresponding mark file of second audio file, wherein according to the state of voice plate to be measured, by institute
It states voice plate to be measured and plays the audio test file, executing test assignment includes：
In the case where the state of the voice plate to be measured is wake-up states, file is marked according to described in second audio file
Test is identified to the voice plate to be measured, obtains the discrimination of the voice plate to be measured.
8. the method according to the description of claim 7 is characterized in that according to file is marked described in second audio file to institute
It states voice plate to be measured and test is identified, the discrimination for obtaining the voice plate to be measured includes：
The instruction word in second audio file is played to the voice plate to be measured by Hi-Fi sound-box；
Obtain the word content that the voice plate to be measured returns；
Judge whether the word content and the information shown of mark document are identical；
If the word content is identical as the information that the mark document is shown, determination identifies successfully；
According to successful number is identified, the discrimination is obtained.
9. according to the method described in claim 8, it is characterized in that, the method further includes：
If the information that the word content shows with the mark document differs, described instruction word is played again；
If the word content that the voice plate to be measured again returns to is identical as the information that the mark document is shown, identification is determined
Success；
If the information that the word content that the voice plate to be measured again returns to shows with the mark document differs, determines and know
Do not fail.
10. the method according to any one of claim 7 to 9, which is characterized in that the mark file include it is following at least
One of：Audio file name, voice content, format.
11. a kind of tone testing device, which is characterized in that including：
Acquiring unit, for obtaining audio test file；
Test cell, for the state according to voice plate to be measured, by playing the audio-frequency test text to the voice plate to be measured
Part executes test assignment；
Generation unit, for generating the test result for being used to indicate the voice plate performance to be measured.
12. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein described program right of execution
Profit requires the tone testing method described in any one of 1 to 10.
13. a kind of terminal, which is characterized in that including：
Memory, for storing program；
Processor, for running described program, wherein perform claim requires any one of 1 to 10 institute when described program is run
The tone testing method stated.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810188414.XA CN108597494A (en) | 2018-03-07 | 2018-03-07 | Tone testing method and device |
| PCT/CN2018/118976 WO2019169914A1 (en) | 2018-03-07 | 2018-12-03 | Method and device for voice testing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810188414.XA CN108597494A (en) | 2018-03-07 | 2018-03-07 | Tone testing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108597494A true CN108597494A (en) | 2018-09-28 |
Family
ID=63625785
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810188414.XA Pending CN108597494A (en) | 2018-03-07 | 2018-03-07 | Tone testing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108597494A (en) |
| WO (1) | WO2019169914A1 (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712608A (en) * | 2019-02-28 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Multitone area wake-up test method, apparatus and storage medium |
| CN109817219A (en) * | 2019-03-19 | 2019-05-28 | 四川长虹电器股份有限公司 | Voice wake-up test method and system |
| WO2019169914A1 (en) * | 2018-03-07 | 2019-09-12 | 珠海格力电器股份有限公司 | Method and device for voice testing |
| CN111179907A (en) * | 2019-12-31 | 2020-05-19 | 深圳Tcl新技术有限公司 | Voice recognition test method, device, equipment and computer readable storage medium |
| CN111341296A (en) * | 2020-02-17 | 2020-06-26 | 诚迈科技（南京）股份有限公司 | Voice-controlled response test method, test machine and storage medium |
| 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 |
| CN111611169A (en) * | 2020-05-22 | 2020-09-01 | 深圳市亿道数码技术有限公司 | Automatic testing method and testing tool for voice assistant wake-up rate |
| CN111739513A (en) * | 2020-07-22 | 2020-10-02 | 江苏清微智能科技有限公司 | Automatic voice awakening test system and test method thereof |
| CN111739512A (en) * | 2020-06-18 | 2020-10-02 | 中汽院智能网联科技有限公司 | Voice wake-up rate testing method, system, device and medium based on real vehicle |
| CN111933108A (en) * | 2020-09-25 | 2020-11-13 | 蘑菇车联信息科技有限公司 | Automatic testing method for intelligent voice interaction system of intelligent network terminal |
| CN111953764A (en) * | 2020-08-07 | 2020-11-17 | 杭州国芯科技股份有限公司 | Artificial intelligence speech algorithm automatic testing method |
| CN112309430A (en) * | 2019-07-31 | 2021-02-02 | 广东美的制冷设备有限公司 | Household appliance and self-checking method and device thereof |
| CN113362806A (en) * | 2020-03-02 | 2021-09-07 | 北京奇虎科技有限公司 | Intelligent sound evaluation method, system, storage medium and computer equipment thereof |
| CN113556661A (en) * | 2020-04-26 | 2021-10-26 | 阿里巴巴集团控股有限公司 | Electroacoustic detection and information output method, system, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102723080A (en) * | 2012-06-25 | 2012-10-10 | 惠州市德赛西威汽车电子有限公司 | Voice recognition test system and voice recognition test method |
| CN104464755A (en) * | 2014-12-02 | 2015-03-25 | 科大讯飞股份有限公司 | Voice evaluation method and device |
| CN106228986A (en) * | 2016-07-26 | 2016-12-14 | 北京奇虎科技有限公司 | The automated testing method of a kind of speech recognition engine, device and system |
| CN107221341A (en) * | 2017-06-06 | 2017-09-29 | 北京云知声信息技术有限公司 | A kind of tone testing method and device |
| CN107516510A (en) * | 2017-07-05 | 2017-12-26 | 百度在线网络技术（北京）有限公司 | A kind of smart machine automated voice method of testing and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103578463B (en) * | 2012-07-27 | 2017-12-01 | 腾讯科技（深圳）有限公司 | Automated testing method and test device |
| CN104837010B (en) * | 2015-04-24 | 2017-04-12 | 青岛海信电器股份有限公司 | Voice remote control testing method, device and system |
| CN108597494A (en) * | 2018-03-07 | 2018-09-28 | 珠海格力电器股份有限公司 | Tone testing method and device |
-
2018
- 2018-03-07 CN CN201810188414.XA patent/CN108597494A/en active Pending
- 2018-12-03 WO PCT/CN2018/118976 patent/WO2019169914A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102723080A (en) * | 2012-06-25 | 2012-10-10 | 惠州市德赛西威汽车电子有限公司 | Voice recognition test system and voice recognition test method |
| CN104464755A (en) * | 2014-12-02 | 2015-03-25 | 科大讯飞股份有限公司 | Voice evaluation method and device |
| CN106228986A (en) * | 2016-07-26 | 2016-12-14 | 北京奇虎科技有限公司 | The automated testing method of a kind of speech recognition engine, device and system |
| CN107221341A (en) * | 2017-06-06 | 2017-09-29 | 北京云知声信息技术有限公司 | A kind of tone testing method and device |
| CN107516510A (en) * | 2017-07-05 | 2017-12-26 | 百度在线网络技术（北京）有限公司 | A kind of smart machine automated voice method of testing and device |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019169914A1 (en) * | 2018-03-07 | 2019-09-12 | 珠海格力电器股份有限公司 | Method and device for voice testing |
| 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 |
| CN109712608A (en) * | 2019-02-28 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Multitone area wake-up test method, apparatus and storage medium |
| CN109817219A (en) * | 2019-03-19 | 2019-05-28 | 四川长虹电器股份有限公司 | Voice wake-up test method and system |
| CN112309430A (en) * | 2019-07-31 | 2021-02-02 | 广东美的制冷设备有限公司 | Household appliance and self-checking method and device thereof |
| CN111179907A (en) * | 2019-12-31 | 2020-05-19 | 深圳Tcl新技术有限公司 | Voice recognition test method, device, equipment and computer readable storage medium |
| CN111341296A (en) * | 2020-02-17 | 2020-06-26 | 诚迈科技（南京）股份有限公司 | Voice-controlled response test method, test machine and storage medium |
| CN111341296B (en) * | 2020-02-17 | 2023-12-12 | 智达诚远科技有限公司 | Voice control response test method, tester and storage medium |
| CN113362806A (en) * | 2020-03-02 | 2021-09-07 | 北京奇虎科技有限公司 | Intelligent sound evaluation method, system, storage medium and computer equipment thereof |
| CN113556661A (en) * | 2020-04-26 | 2021-10-26 | 阿里巴巴集团控股有限公司 | Electroacoustic detection and information output method, system, equipment and storage medium |
| CN111611169A (en) * | 2020-05-22 | 2020-09-01 | 深圳市亿道数码技术有限公司 | Automatic testing method and testing tool for voice assistant wake-up rate |
| CN111739512A (en) * | 2020-06-18 | 2020-10-02 | 中汽院智能网联科技有限公司 | Voice wake-up rate testing method, system, device and medium based on real vehicle |
| CN111739513A (en) * | 2020-07-22 | 2020-10-02 | 江苏清微智能科技有限公司 | Automatic voice awakening test system and test method thereof |
| CN111953764A (en) * | 2020-08-07 | 2020-11-17 | 杭州国芯科技股份有限公司 | Artificial intelligence speech algorithm automatic testing method |
| CN111953764B (en) * | 2020-08-07 | 2023-04-07 | 杭州国芯科技股份有限公司 | Artificial intelligence speech algorithm automatic testing method |
| CN111933108A (en) * | 2020-09-25 | 2020-11-13 | 蘑菇车联信息科技有限公司 | Automatic testing method for intelligent voice interaction system of intelligent network terminal |
| CN111933108B (en) * | 2020-09-25 | 2021-01-12 | 蘑菇车联信息科技有限公司 | Automatic testing method for intelligent voice interaction system of intelligent network terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019169914A1 (en) | 2019-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108597494A (en) | Tone testing method and device | |
| USRE44248E1 (en) | System for transferring personalize matter from one computer to another | |
| CN111798833B (en) | Voice test method, device, equipment and storage medium | |
| Friederici | Language comprehension: A biological perspective | |
| CN107103915A (en) | A kind of audio data processing method and device | |
| CN107516510A (en) | A kind of smart machine automated voice method of testing and device | |
| CN107086040A (en) | Speech recognition capabilities method of testing and device | |
| CN109003602A (en) | Test method, device, equipment and the computer-readable medium of speech production | |
| CN109256115A (en) | A kind of speech detection system and method for intelligent appliance | |
| CN110473525A (en) | The method and apparatus for obtaining voice training sample | |
| CN110379410A (en) | Voice response speed automatic analysis method and system | |
| CN102609406B (en) | Learning device, judgment means, learning method and determination methods | |
| CN110197658A (en) | Method of speech processing, device and electronic equipment | |
| CN109726105A (en) | Test data building method, device, equipment and storage medium | |
| US20110144990A1 (en) | Rating speech naturalness of speech utterances based on a plurality of human testers | |
| CN109325091A (en) | Update method, device, equipment and the medium of points of interest attribute information | |
| CN110491369A (en) | Appraisal procedure, device, storage medium and the electronic equipment of spoken grade | |
| CN104505103B (en) | Voice quality assessment equipment, method and system | |
| CN110070847A (en) | Musical sound assessment method and Related product | |
| CN109493869A (en) | The acquisition method and system of audio data | |
| CN106601220A (en) | Method and device for recording antiphonal singing of multiple persons | |
| CN109712608A (en) | Multitone area wake-up test method, apparatus and storage medium | |
| CN107767850A (en) | A kind of singing marking method and system | |
| CN108597538A (en) | The evaluating method and system of speech synthesis system | |
| CN110503960A (en) | Uploaded in real time method, apparatus, equipment and the storage medium of speech recognition result |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |
|
| RJ01 | Rejection of invention patent application after publication |