CN204440902U - Terminal intelligent tone testing system - Google Patents
Terminal intelligent tone testing system Download PDFInfo
- Publication number
- CN204440902U CN204440902U CN201420820284.4U CN201420820284U CN204440902U CN 204440902 U CN204440902 U CN 204440902U CN 201420820284 U CN201420820284 U CN 201420820284U CN 204440902 U CN204440902 U CN 204440902U
- Authority
- CN
- China
- Prior art keywords
- terminal
- model
- measured
- intelligent sound
- test
- 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.)
- Expired - Fee Related
Links
Abstract
The utility model discloses terminal intelligent tone testing system, comprise terminal to be measured, test computer and intelligent sound engine or model, wherein, intelligent sound engine or model are located at high in the clouds or terminal local to be measured, test computer is connected with terminal to be measured, and terminal to be measured is wirelessly connected with high in the clouds intelligent sound engine or model.Intelligent sound engine or model beyond the clouds when, terminal to be measured wirelessly connects high in the clouds and identifies, intelligent sound engine or model are when this locality, terminal to be measured directly identifies in this locality, in both cases, test computer is all direct obtains recognition result from terminal to be measured, carries out performance evaluation.Compared with prior art, the utility model can test the intelligent sound performance in terminal to be measured self and high in the clouds, and test is comprehensive, accurate, and simple and flexible is easy-to-use.
Description
Technical field
The utility model relates to a kind of test macro of terminal, and especially the intelligent sound test macro of terminal, belongs to automatization testing technique field.
Background technology
The mode of the intelligent sound of current test terminal is: on test computer, install intelligent sound engine or model, the speech data that terminal to be measured is enrolled is passed to intelligent sound engine on test computer or model couples together, and identifies.There is following shortcoming in this kind of test mode:
1, the engine of certain intelligent sound or the performance of model can only be tested;
2, to can not get in measured terminal or its engine of concrete intelligence voice beyond the clouds or the performance of model;
3, not comprehensive, situation about also using with reality has difference.
Modern terminal, as mobile phone etc., comprises following several form with the connection of intelligent sound:
1) the intelligent sound engine in terminal or model, namely after terminal to be measured obtains the speech data of user, directly pass to self with intelligent sound engine or model, carry out intelligent sound identification other, provide result, the engine of the intelligent sound of various different terminals or model are different;
2) in terminal with intelligent sound front-end processing, and the engine of intelligent sound or model are in its high in the clouds, terminal is just the data after front-end processing, as phonetic feature, pass to its high in the clouds to identify, provide result, intelligent sound engine or the model in different its high in the clouds of intelligent sound software are also different;
3) terminal is by communication mode, as made a phone call, to certain server, carries out intelligent sound identification, the speech data being through codec that terminal to be measured like this sends.
Above-mentioned intelligent sound comprises speech recognition, voice wake up and Speaker Identification etc., and the intelligent sound test of communication terminal is mainly assessed for the local intelligent voice of terminal and the performance of self high in the clouds intelligent sound and other servers or high in the clouds intelligent sound.Not yet there is effective automatization test system at present to realize the intelligent sound test of terminal local or high in the clouds.
Utility model content
Goal of the invention: for the deficiencies in the prior art, the utility model object is the automatization test system providing communication terminal intelligent sound, to realize the test of communication terminal local intelligent voice or high in the clouds intelligent sound.
Technical scheme: test for realizing the intelligent sound of communication terminal in its high in the clouds, terminal intelligent tone testing system of the present utility model, comprise terminal to be measured, test computer and intelligent sound engine or model, described intelligent sound engine or model are located at high in the clouds, test computer is connected with terminal to be measured, and terminal to be measured is wirelessly connected with high in the clouds intelligent sound engine or model.
Preferably, described wireless mode is Wifi or bluetooth or other similar fashion, does not need to spend flow during test.
Preferably, described system also comprises speech player and noise generating device, and speech player is connected with test computer.Noise generating device, for generation of the noise of test environment, can be used for testing the recognition performance under different noise environment.The position relationship of speech player and noise generating device and terminal to be measured can be determined according to actual testing requirement.
Preferably, described terminal to be measured comprises test interface module and the voice acquisition module be connected successively, intelligent sound front end and recognition result output module, described test interface module is connected with recognition result output module, described test interface module is connected with test computer, and described intelligent sound front end is wirelessly connected with its high in the clouds intelligent sound engine or model.
For realizing the intelligent sound test of communication terminal this locality, terminal intelligent tone testing system of the present utility model, comprise terminal to be measured, test computer and intelligent sound engine or model, described intelligent sound engine or model are located at terminal local to be measured, and test computer is connected with terminal to be measured.
Preferably, described system also comprises speech player and noise generating device, and speech player is connected with test computer.Noise generating device, for generation of the noise of test environment, can be used for testing the recognition performance under different noise environment.The position relationship of speech player and noise generating device and terminal to be measured can be determined according to actual testing requirement.
Preferably, described terminal to be measured comprises test interface module and the voice acquisition module be connected successively, intelligent sound engine or model and recognition result output module, described test interface module is connected with recognition result output module, and described test interface module is connected with test computer.
Beneficial effect: the terminal intelligent tone testing system that the utility model provides, intelligent sound engine or model are arranged on terminal local to be measured or its high in the clouds, do not need various intelligent sound engine or model are installed on test computer, more without the need to transmitting the speech data of admission between terminal to be measured and test computer, in terminal, directly carry out intelligent sound identification or connect its high in the clouds by terminal identifying.When intelligent sound engine or model beyond the clouds, terminal is wirelessly connected with high in the clouds, as Wifi, do not need flow like this, and do not need the interface understanding terminal and high in the clouds (to be transferring voice feature? or directly transferring voice file? or other).Further, test macro directly obtains the recognition result of intelligent sound from terminal by the interface with terminal, take full advantage of the function of terminal to be measured self, simplify system, and be applicable to intelligent sound engine or model in two kinds of situations local and beyond the clouds.Compared with prior art, the utility model had both ensured test automation and ease for use, reduced system complexity and cost, raising testing efficiency again, and test more comprehensively, more accurate.
Accompanying drawing explanation
Fig. 1 is the structural representation of the utility model embodiment 1;
Fig. 2 is the structural representation of the utility model embodiment 2;
Fig. 3 is the process flow diagram of the method for embodiment 1;
Fig. 4 is the process flow diagram of the method for embodiment 2.
Embodiment
Below in conjunction with specific embodiment, illustrate the utility model further.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combination in any mutually.Should understand, these embodiments are only not used in restriction scope of the present utility model for illustration of the utility model, after having read the utility model, the amendment of those skilled in the art to the various equivalent form of value of the present utility model has all fallen within the application's claims item and has required limited range.
Embodiment 1:
As shown in Figure 1, the terminal intelligent tone testing system that the present embodiment provides, comprise terminal to be measured, test computer and intelligent sound engine or model, wherein intelligent sound engine or model are located at terminal local to be measured, and test computer is connected with terminal to be measured.Terminal to be measured also comprises voice acquisition module, recognition result output module and test interface module, voice acquisition module is used for recording, and be connected with local intelligent speech engine or model, import the speech data of admission into intelligent sound engine or model identifies, intelligent sound engine or model output to recognition result output module after obtaining recognition result; Test interface module is connected with test computer and recognition result output module respectively, receive the signal of test computer and obtain recognition result to test computer from recognition result output module, test interface module also can be connected with voice acquisition module or intelligent sound engine or model, to assign the testing and control instruction of test computer.Test computer can play voice messaging by the loudspeaker carried, and also can carry out broadcasting voice by external speech player, speech player can be general loudspeaker or special playback equipment, as artificial mouth, to adapt to various different test request.
Method of testing for embodiment 1 mainly comprises: test computer controls to play the voice messaging preset; Terminal to be measured connects local intelligent speech engine or model identifies the speech data of enrolling, and exports recognition result in terminal to be measured; Test computer obtains recognition result and carries out performance evaluation based on recognition result.As shown in Figure 3, the detailed step of method of testing comprises:
1) test computer sends the signal starting test to test interface module, and test interface module controls voice acquisition module and enters local recording state;
2) test computer exports voice messaging corresponding to referenced text preset to speech player, and speech player plays voice;
3) after speech play terminates, test computer sends the signal play and terminate to test interface module, and test interface module controls voice acquisition module and exits local recording state;
4) terminal startup local intelligent speech engine or model carry out intelligent sound identification to the speech data of enrolling;
5) test computer is inquired about terminal to be measured by test interface module and whether is obtained voice identification result, if obtain recognition result, read resulting text, otherwise inquire about terminal to be measured again every the time interval (as 100ms) of setting and whether obtain voice identification result until obtain recognition result or exceed the total duration of wait (as 1s) of setting;
Can compare with referenced text after test computer obtains voice identification result text, the calculating of discrimination can be carried out based on repeatedly test result, carry out performance evaluation and the assessment of intelligent sound identification further.
Embodiment 2:
As shown in Figure 2, the terminal intelligent tone testing system that the present embodiment provides, comprise terminal to be measured, test computer and intelligent sound engine or model, intelligent sound engine or model arrange high in the clouds, terminal to be measured is provided with intelligent sound front end, terminal to be measured wirelessly, as Wifi, bluetooth or other wireless modes, be connected with high in the clouds intelligent sound engine or model, test computer is connected with terminal to be measured.Terminal to be measured also comprises voice acquisition module, recognition result output module and test interface module, voice acquisition module is used for recording, and be connected with local intelligent speech front-end, import the speech data of admission into intelligent sound front end, front end wirelessly, as Wifi, is connected with high in the clouds intelligent sound engine or model, recognition result is wirelessly passed back by high in the clouds, outputs to recognition result output module; Test interface module is connected with test computer and recognition result output module respectively, receive the signal of test computer and obtain recognition result to test computer from recognition result output module, test interface module also can be connected with voice acquisition module or intelligent sound front end, to assign the testing and control instruction of test computer.The external speech player of test computer carries out broadcasting voice, and test computer can connect noise generating device, controls the broadcasting of test environment noise.It should be understood that the noise of noise generating device for generation of test environment, can independently control, also can be connected with test computer, the control signal of the broadcasting that receiving computer carries out in testing process, stopping, volume, noise source type etc.
Method of testing for embodiment 2 mainly comprises: test computer controls to play the voice messaging preset; Terminal to be measured wirelessly connects high in the clouds intelligent sound engine or model identifies the speech data of enrolling, and exports recognition result in terminal to be measured; Test computer obtains recognition result and carries out performance evaluation based on recognition result.As shown in Figure 4, the detailed step of method of testing comprises:
1) test computer control noises generation device starts to play noise;
2) test computer sends the signal starting test to test interface module, and test interface module controls voice acquisition module and enters local recording state;
3) test computer exports voice messaging corresponding to referenced text preset to speech player, and speech player plays voice;
4) after speech play terminates, test computer sends the signal play and terminate to test interface module, and test interface module controls voice acquisition module and exits local recording state;
5) test computer control noises generation device stops playing noise;
6) after the speech data of enrolling is carried out intelligent sound front-end processing by terminal to be measured, wirelessly (as Wifi) passes to intelligent sound engine or the model in high in the clouds, startup high in the clouds intelligent sound engine or model carry out intelligent sound identification, and the recognition result that wait-receiving mode is wirelessly passed back;
7) test computer is inquired about terminal to be measured by test interface module and whether is obtained voice identification result, if obtain recognition result, read resulting text, otherwise inquire about terminal to be measured again every the time interval (as 100ms) of setting and whether obtain voice identification result until obtain recognition result or exceed the total duration of wait (as 4s) of setting.
Can compare with referenced text after test computer obtains voice identification result text, the calculating of discrimination can be carried out based on repeatedly test result, carry out performance evaluation and the assessment of intelligent sound identification further.
Claims (7)
1. terminal intelligent tone testing system, comprise terminal to be measured, test computer and intelligent sound engine or model, it is characterized in that, described intelligent sound engine or model are located at high in the clouds, test computer is connected with terminal to be measured, and terminal to be measured is wirelessly connected with high in the clouds intelligent sound engine or model.
2. terminal intelligent tone testing system according to claim 1, is characterized in that, described wireless mode is Wifi or bluetooth.
3. terminal intelligent tone testing system according to claim 1 and 2, it is characterized in that, described system also comprises speech player and noise generating device, and speech player is connected with test computer.
4. terminal intelligent tone testing system according to claim 1 and 2, it is characterized in that, described terminal to be measured comprises test interface module and the voice acquisition module be connected successively, intelligent sound front end and recognition result output module, described test interface module is connected with recognition result output module, described test interface module is connected with test computer, and described intelligent sound front end is wirelessly connected with its high in the clouds intelligent sound engine or model.
5. terminal intelligent tone testing system, comprises terminal to be measured, test computer and intelligent sound engine or model, it is characterized in that, described intelligent sound engine or model are located at terminal local to be measured, and test computer is connected with terminal to be measured.
6. terminal intelligent tone testing system according to claim 5, it is characterized in that, described system also comprises speech player and noise generating device, and speech player is connected with test computer.
7. terminal intelligent tone testing system according to claim 5, it is characterized in that, described terminal to be measured comprises test interface module and the voice acquisition module be connected successively, intelligent sound engine or model and recognition result output module, described test interface module is connected with recognition result output module, and described test interface module is connected with test computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420820284.4U CN204440902U (en) | 2014-12-22 | 2014-12-22 | Terminal intelligent tone testing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201420820284.4U CN204440902U (en) | 2014-12-22 | 2014-12-22 | Terminal intelligent tone testing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN204440902U true CN204440902U (en) | 2015-07-01 |
Family
ID=53608737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201420820284.4U Expired - Fee Related CN204440902U (en) | 2014-12-22 | 2014-12-22 | Terminal intelligent tone testing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN204440902U (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106332162A (en) * | 2015-06-25 | 2017-01-11 | 中兴通讯股份有限公司 | Telephone traffic test system and method |
| CN108665900A (en) * | 2018-04-23 | 2018-10-16 | 百度在线网络技术（北京）有限公司 | High in the clouds awakening method and system, terminal and computer readable storage medium |
| US10964317B2 (en) | 2017-07-05 | 2021-03-30 | Baidu Online Network Technology (Beijing) Co., Ltd. | Voice wakeup method, apparatus and system, cloud server and readable medium |
-
2014
- 2014-12-22 CN CN201420820284.4U patent/CN204440902U/en not_active Expired - Fee Related
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106332162A (en) * | 2015-06-25 | 2017-01-11 | 中兴通讯股份有限公司 | Telephone traffic test system and method |
| US10964317B2 (en) | 2017-07-05 | 2021-03-30 | Baidu Online Network Technology (Beijing) Co., Ltd. | Voice wakeup method, apparatus and system, cloud server and readable medium |
| CN108665900A (en) * | 2018-04-23 | 2018-10-16 | 百度在线网络技术（北京）有限公司 | High in the clouds awakening method and system, terminal and computer readable storage medium |
| CN108665900B (en) * | 2018-04-23 | 2020-03-03 | 百度在线网络技术（北京）有限公司 | Cloud wake-up method and system, terminal and computer readable storage medium |
| US11574632B2 (en) | 2018-04-23 | 2023-02-07 | Baidu Online Network Technology (Beijing) Co., Ltd. | In-cloud wake-up method and system, terminal and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104538042A (en) | Intelligent voice test system and method for terminal | |
| CN112863547B (en) | Virtual resource transfer processing method, device, storage medium and computer equipment | |
| CN104580699B (en) | Acoustic control intelligent terminal method and device when a kind of standby | |
| CN111341296B (en) | Voice control response test method, tester and storage medium | |
| CN108428452A (en) | Terminal support and far field voice interactive system | |
| US20180293982A1 (en) | Voice assistant extension device and working method therefor | |
| US20210241768A1 (en) | Portable audio device with voice capabilities | |
| CN109473095A (en) | A kind of intelligent home control system and control method | |
| CN109326305B (en) | Method and system for batch testing of speech recognition and text synthesis | |
| CN204440902U (en) | Terminal intelligent tone testing system | |
| WO2017071183A1 (en) | Voice processing method and device, and pickup circuit | |
| CN109147761A (en) | Test method based on batch speech recognition and TTS text synthesis | |
| KR102512614B1 (en) | Electronic device audio enhancement and method thereof | |
| CN110738994A (en) | Control method, device, robot and system for smart homes | |
| CN110675848A (en) | Audio processing method, device and storage medium | |
| CN109510636A (en) | Receive test method, device, medium and the equipment of FM signal function | |
| CN209184746U (en) | Microphone system | |
| CN108712218A (en) | A method of detection simulation talk back equipment closely utter long and high-pitched sounds possibility by call | |
| CN111933137A (en) | Voice wake-up test method and device, computer readable medium and electronic device | |
| CN113053371A (en) | Voice control system and method, voice suite, bone conduction and voice processing device | |
| CN106094673A (en) | Automobile wakes up word system and control method thereof up | |
| CN208541311U (en) | Hearing test apparatus | |
| CN110660376A (en) | Audio processing method, device and storage medium | |
| CN206756741U (en) | A kind of blood glucose meter based on the communication of mobile device Lightning mouths | |
| CN206249889U (en) | A kind of speech recognition evaluating apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150701 Termination date: 20181222 |