Summary of the invention
The problem that the present invention solves is to provide a kind of pronunciation inputting method and voice entry system thereof of mobile unit of the voice messaging that obtains user that can be complete.
For addressing the above problem, embodiments of the invention provide a kind of pronunciation inputting method of mobile unit, comprising:
Start speech voice input function;
After starting speech voice input function, opening voice receiving trap, and when opening voice receiving trap or after opening voice receiving trap, prompting user input voice;
After prompting user input voice, input voice information is to pronunciation receiver;
After completing the input of voice messaging, the ground unrest that filtering pronunciation receiver receives;
Voice messaging in identification pronunciation receiver after filter out background noise.
Alternatively, the method for prompting user input voice is: play and start to indicate sound.
Alternatively, the described frequency that starts to indicate sound is less than 300 hertz, or is greater than 3400 hertz, and the described frequency that starts to indicate sound immobilizes.
The method of the ground unrest that filtering pronunciation receiver alternatively, receives is: obtain the voice messaging that pronunciation receiver receives; By ground unrest elimination algorithm, the information that described voice messaging medium frequency is identical with the described frequency that starts to indicate sound is removed.
Alternatively, also comprise: after starting speech voice input function, whether voice message user wants help, then opening voice receiving trap.
Alternatively, also comprise: after the voice messaging in identification pronunciation receiver after filter out background noise, obtain user's demand; Start with user's demand and serve accordingly.
Inventor also provides a kind of voice entry system of mobile unit, comprising:
Start voice unit, for starting speech voice input function;
Voice receiving unit, for opening after starting speech voice input function, receives the voice messaging of input;
Input prompt unit, while opening for voice receiving unit or after opening, prompting user input voice;
Audio Processing Unit, the voice messaging receiving for the treatment of voice receiving unit, the ground unrest that filtering receives;
Voice recognition unit, the voice messaging after identification Audio Processing Unit filter out background noise.
Alternatively, described input prompt unit comprises the first player, for playing the unlatching of fixed frequency, indicates sound, prompting user input voice.
Alternatively, described Audio Processing Unit comprises: the first acquiring unit, the voice messaging receiving for obtaining voice receiving unit; Noise-filtering unit, for filter out background noise.
Alternatively, also comprise: voice alerting unit, for starting after speech voice input function, whether voice message user wants help, and described voice receiving unit is opened after voice message completes.
Alternatively, also comprise: second acquisition unit, for receiving the voice messaging of voice recognition unit identification, obtains user's demand; Service start unit, serves accordingly for starting with user's demand.
Compared with prior art, technical scheme of the present invention has the following advantages:
When opening voice receiving trap or after opening voice receiving trap, prompting user input voice, user loquiturs after described prompting obtaining, and has avoided user to rob the information that words cause pronunciation receiver to receive imperfect.And after the input that completes voice messaging until user, filtering the ground unrest that receives of pronunciation receiver, eliminated the interference that ground unrest causes follow-up identification voice messaging, the voice messaging recognizing is clear complete.
Further, after starting speech voice input function, whether voice message user wants help, then opening voice receiving trap.Increase the interaction between user and mobile unit, designed hommization more.
Further, after ground unrest in the voice messaging receiving in filtering, the voice messaging of identification user input that can be accurately complete, clearly obtain user's demand, then provide and serve accurately to user, the process that user uses described mobile unit to serve is more smooth and easy, convenient.
In the voice entry system of mobile unit of the present invention, input prompt unit is when voice receiving unit is opened or after opening, prompting user input voice, and the ground unrest in the voice messaging that Audio Processing Unit meeting filtering voice receiving unit receives, described voice entry system simple in structure, the voice messaging recognizing is clear complete, and being convenient to follow-uply provides service more accurately for user, facilitate user, improved the security of user's driving procedure.
Embodiment
As described in background, the pronunciation inputting method of the mobile unit of prior art, often exists user to rob the situation of words, cause mobile unit not can completely obtain user's voice messaging, and cause speech recognition failure, cannot realize user's demand.
Through research, find, the main cause that prior art generation user robs words is: user cannot accurately learn that mobile unit starts the time of typing user speech.If mobile unit is before typing user's voice messaging, give user prompting, inform that user starts recording now, user carries out the input of voice messaging after obtaining above-mentioned prompting again, mobile unit is can completely typing user's voice messaging, and subsequent speech recognition success ratio is increased.Therefore, inventor provides a kind of pronunciation inputting method of mobile unit, after opening the speech voice input function of mobile unit, when opening sound-track engraving apparatus, or play and start to indicate sound after unlatching sound-track engraving apparatus, for example " drip ", make user hear " dripping " start after indicating sound start again input voice information.
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.
Please refer to Fig. 1, Fig. 1 shows the pronunciation inputting method of the mobile unit of first embodiment of the invention, comprising:
Step S101, starts speech voice input function;
Step S103, after starting speech voice input function, opening voice receiving trap, and when opening voice receiving trap or after opening voice receiving trap, prompting user input voice;
Step S105, after prompting user input voice, input voice information is to pronunciation receiver;
Step S107, after completing the input of voice messaging, the ground unrest that filtering pronunciation receiver receives;
Step S109, the voice messaging in identification pronunciation receiver after filter out background noise;
Step S111, after the voice messaging in identification pronunciation receiver after filter out background noise, obtains user's demand;
Step S113, starts with user's demand and serves accordingly.
Wherein, the method for described startup speech voice input function comprises: start the phonetic entry button of mobile unit end, or select by the icon on the touch display screen of mobile unit end.In an embodiment of the present invention, for improving user's experience, increase the security performance of user's driving procedure, be provided with phonetic entry button on the bearing circle of described vehicle, user presses described phonetic entry button, has started speech voice input function.
Described pronunciation receiver receiving speech information, user can typing demand for services by pronunciation receiver.In an embodiment of the present invention, described pronunciation receiver is sound-track engraving apparatus, after starting speech voice input function, opens sound-track engraving apparatus.
Inventor finds, starts speech voice input function poor to the unlatching life period of pronunciation receiver, starts after speech voice input function, and pronunciation receiver needs certain reaction time to open.If user just loquiturs within the described mistiming, the incomplete situation of voice messaging that can exist pronunciation receiver to receive.For example, user starts speech voice input function, after 3 seconds, pronunciation receiver is just opened, and user just loquiturs after starting speech voice input function for 1 second: " I will send short messages to Li little Lei ", so, the actual user's who receives of pronunciation receiver voice messaging is " Li little Lei ", causes the voice messaging of reception imperfect, affect and follow-uply obtain user's request, and have influence on the quality of service-user.
Further, inventor finds, if when pronunciation receiver is opened or after opening, mobile unit can give user prompting, informs that user speech receiving trap opens, can receiving speech information, prompting user input voice, can avoid the generation of above-mentioned situation.
In an embodiment of the present invention, for making to point out the method for user input voice short and sweet, the method of described prompting user input voice is: play and to start to indicate sound, user can start input voice information hearing after starting to indicate sound, and the voice messaging that pronunciation receiver receives is complete.For example, when opening sound-track engraving apparatus, broadcasting starts to indicate sound " drips ", informs that user recording machine has started recording, can receive the voice messaging of user's input, user starts to indicate after sound what hear " dripping ", just start speech " I will send short messages to Li little Lei ", the voice messaging that sound-track engraving apparatus receives is complete, is " I will send short messages to Li little Lei ", be beneficial to the follow-up user's request of obtaining, then corresponding service is provided.
While opening due to pronunciation receiver or after opening, given user speech prompting, the voice messaging that described pronunciation receiver receives, except the voice that comprise user's input, also comprises information of voice prompt.And, the noise of external environment, for example the sound of blowing a whistle of vehicle also may be received by pronunciation receiver.During subsequent speech recognition, the voice messaging that real Useful Information is user's input.Therefore, the voice messaging that need to receive pronunciation receiver carries out filtering, and filter out background noise, such as noise of information of voice prompt and external environment etc.
Although the frequency that each user speaks is not identical, but the frequency of speaking due to user is all between 300 hertz-3400 hertz, as long as the frequency of prompting user input voice outside above-mentioned scope, can guarantee the complete of user speech information after filter out background noise.In an embodiment of the present invention, the described frequency that starts to indicate sound is less than 300 hertz, or be greater than 3400 hertz, and the described frequency that starts to indicate sound is when immobilize, and the method for the ground unrest of filtering pronunciation receiver is: obtain the voice messaging that pronunciation receiver receives; By ground unrest elimination algorithm, the information that described voice messaging medium frequency is identical with the described frequency that starts to indicate sound is removed.The user speech information that in this, method of filter out background noise can't receive pronunciation receiver impacts, and can guarantee the integrality of the voice messaging that user inputs.
Because user is after mobile unit prompting user input voice, just input voice to pronunciation receiver, and the user speech information pronunciation receiver not being received during filter out background noise impacts, therefore, after the voice messaging that mobile unit receives in identification pronunciation receiver, can obtain accurately user's demand.In embodiments of the invention, the voice messaging recognizing is " I will send note to Li little Lei ", and user's request is " transmission note ", and object is " Li little Lei ".
Mobile unit, after obtaining user's request, can start with user's request and serve accordingly.For example, in embodiments of the invention, mobile unit gets after the demand of customer requirements " transmission note ", starts note transmitting element, and recalls note transmission object " Li little Lei ".As long as user inputs short message content, mobile unit can send to Li little Lei by short message content automatically.Whole process is simple, convenient, workable, and has greatly improved the security that user drives.
It should be noted that, in other embodiments of the invention, after starting speech voice input function, whether voice message user wants help, then opening voice receiving trap.For example, user starts after the phonetic entry button of mobile unit end, and mobile unit can voice message user " boss be may I ask you and wanted help ", and after the complete user of voice message 1 second, play and start to indicate sound, prompting user can start to input voice.This process is owing to having increased voice message user's step, and making has had warm interaction between user and mobile unit, seems humanized.
In embodiments of the invention, due in opening voice receiving trap, or after opening voice receiving trap, prompting user input voice, user can know pronunciation receiver start recording time, avoided user rob words cause pronunciation receiver to receive voice messaging imperfect, and follow-up in the situation that do not damage the voice messaging receiving, filtering ground unrest, improved follow-up accuracy rate while obtaining user's request, can be convenient provide service for user.
Accordingly, please refer to Fig. 2, inventor also provides a kind of voice entry system of mobile unit, comprising:
Start voice unit 201, for starting speech voice input function;
Voice receiving unit 203, for opening after starting speech voice input function, receives the voice messaging of input;
Input prompt unit 205, while opening for voice receiving unit 203 or after opening, prompting user 280 input voice;
Audio Processing Unit 207, the voice messaging receiving for the treatment of voice receiving unit 203, the ground unrest that filtering receives;
Voice recognition unit 209, the voice messaging after identification Audio Processing Unit 207 filter out background noise;
Second acquisition unit 211, for receiving the voice messaging of voice recognition unit 209 identifications, obtains user's demand;
Service start unit 213, serves accordingly for starting with user 280 demand.
Wherein, described startup voice unit 201 comprises: start the phonetic entry button of mobile unit end, or by the icon on the touch display screen of mobile unit end.User can be by pressing the phonetic entry button of mobile unit end or selecting to start speech voice input function by touching the icon of mobile unit end touch display screen.
Described voice receiving unit 203 is connected with described startup voice unit 201, after starting speech voice input function user 280, opens, and receives the voice messaging of user's 280 inputs.In an embodiment of the present invention, described voice receiving unit 203 is pronunciation receiver, for example sound-track engraving apparatus.
Described input prompt unit 205 is connected with described voice receiving unit 203, for when voice receiving unit 203 is opened, or after voice receiving unit 203 is opened, prompting user 280 input voice.In an embodiment of the present invention, described input prompt unit 205 comprises the first player (not shown), for playing the unlatching of fixed frequency, indicates sound, prompting user input voice.Do not destroy the user speech information integrity that voice receiving unit 203 receives when making subsequent voice processing unit 207 filter out background noise, the frequency that described unlatching indicates sound is less than 300 hertz, or is greater than 3400 hertz.And the frequency that described unlatching indicates sound immobilizes, and is beneficial to follow-up filtering.In an embodiment of the present invention, the frequency that described unlatching indicates sound immobilizes, and is 200 hertz.
It should be noted that, because input prompt unit 205 is when voice receiving unit 203 is opened, or after voice receiving unit 203 is opened, can prompting user 280 input voice, user 280 just inputs voice after receiving described prompting, and the voice messaging that voice receiving unit 203 receives is more complete.For example, in embodiments of the invention, the voice messaging that described voice receiving unit 203 receives is " drip, I will send note to Li little Lei ".
Described Audio Processing Unit 207 is connected with described voice receiving unit 203, for the voice messaging that voice receiving unit 203 is received, process, filtering voice receiving unit 203 is when receiving the voice messaging of user's 280 inputs, the ground unrest simultaneously receiving, the prompting user's 280 that for example noise of external environment and input prompt unit 205 send sound.Audio Processing Unit 207 comprises: the first acquiring unit 2071, the voice messaging receiving for obtaining voice receiving unit 203; Noise-filtering unit 2072, the ground unrest of the voice messaging obtaining for filtering the first acquiring unit 2071.
In an embodiment of the present invention, the method of the ground unrest in the voice messaging that described noise-filtering unit 2072 filtering the first acquiring units 2071 obtain is: by ground unrest elimination algorithm, the information that the voice messaging medium frequency that the first acquiring unit 2071 is obtained is identical with the frequency that starts to indicate sound is removed, for example, by frequency, be " dripping " sound removal of 200 hertz.After the ground unrest that 207 filterings receive through Audio Processing Unit, remaining information is the voice messaging of complete user's input, for example " I will send note to Li little Lei ".
Described voice recognition unit 209 is connected with described Audio Processing Unit 207, for identifying the voice messaging after Audio Processing Unit 207 filter out background noise.In embodiments of the invention, the information that described voice recognition unit 209 recognizes is " I will send note to Li little Lei ".
Described second acquisition unit 211 is connected with voice recognition unit 209, the voice messaging identifying for receiving voice recognition unit 209, and obtain user 280 demand.For example, " sending short messages ", sending object is " Li little Lei ".
Described service start unit 213 is connected with second acquisition unit 211, after receiving user's 280 demands that second acquisition unit 211 obtains, starts with user 280 demand and serves accordingly.For example, in embodiments of the invention, service start unit 213, after receiving the user's request of " sending short messages ", starts note transmitting element, and recalls the communication information that sends object " Li little Lei ".As long as user's 280 inputs (phonetic entry or manually input) short message content, can send to Li little Lei by short message content automatically.In embodiments of the invention, owing to can accurately receiving user 280 demand, therefore serving start unit 213 can start corresponding service accurately, better for user 280 provides service.
It should be noted that, in an embodiment of the present invention, for making interaction warm between user and mobile unit, hommization more, the voice entry system of described mobile unit also comprises: voice alerting unit 215, be used for starting after speech voice input function, whether voice message user wants help, and described voice receiving unit 203 is opened after voice message completes.For example, user, press after the phonetic entry button of mobile unit end, voice message user " boss, may I ask you needs anything to help ", then opening voice receiving element 203, and play and start to indicate sound and " drip ".
To sum up, when opening voice receiving trap or after opening voice receiving trap, prompting user input voice, user loquiturs after described prompting obtaining, and has avoided user to rob the information that words cause pronunciation receiver to receive imperfect.And after the input that completes voice messaging until user, filtering the ground unrest that receives of pronunciation receiver, eliminated the interference that ground unrest causes follow-up identification voice messaging, the voice messaging recognizing is clear complete.
Further, after starting speech voice input function, whether voice message user wants help, then opening voice receiving trap.Increase the interaction between user and mobile unit, designed hommization more.
Further, after ground unrest in the voice messaging receiving in filtering, the voice messaging of identification user input that can be accurately complete, clearly obtain user's demand, then provide and serve accurately to user, the process that user uses described mobile unit to serve is more smooth and easy, convenient.
In the voice entry system of mobile unit of the present invention, input prompt unit is when voice receiving unit is opened or after opening, prompting user input voice, and the ground unrest in the voice messaging that Audio Processing Unit meeting filtering voice receiving unit receives, described voice entry system simple in structure, the voice messaging recognizing is clear complete, and being convenient to follow-uply provides service more accurately for user, facilitate user, improved the security of user's driving procedure.
Although the present invention with preferred embodiment openly as above; but it is not for limiting the present invention; any those skilled in the art without departing from the spirit and scope of the present invention; can utilize method and the technology contents of above-mentioned announcement to make possible change and modification to technical solution of the present invention; therefore; every content that does not depart from technical solution of the present invention; any simple modification, equivalent variations and the modification above embodiment done according to technical spirit of the present invention, all belong to the protection domain of technical solution of the present invention.