Embodiment
In the following description, many details are elaborated to make person of ordinary skill in the field more fullyUnderstand the present invention.But for the technical staff in art it is evident that the realization of the present invention can not have theseSome in detail.However, it should be understood that the present invention is not limited to introduced specific embodiment.On the contrary, can be withConsider to implement the present invention with any combination of following feature and key element, regardless of whether they are related to different embodiments.Therefore, aspect, feature, embodiment and advantage below is used and is not construed as the key element or limit of claim for illustrative purposes onlyIt is fixed, unless clearly proposing in the claims.
In view of fixed voice message mode can not adapt to different scenes in the prior art, the present inventor carriesGo out, before voice message is reported, determine current report scene by detecting in advance, and voice is adjusted based on scene is reportedThe mode of prompting, the defects of prior art should be able to be improved.
Shown in reference picture 1, according to a kind of embodiment of dynamic voice method of the present invention, it includes：
Step s10, report event is formed according to the prompting associated data gathered；
Step s20, is asked according to voice message, it is determined that the voice message mode under the report event；
Step s30, the report of voice message is carried out in a manner of identified voice message.
In above-mentioned embodiment, the characteristic element that scene is reported due to forming has differences according to the difference of scene, thereforeThe data that need to be gathered can accordingly be chosen according to the demand and software and hardware resources of practical application.But generally speaking, determine to reportThe factor of scene it is general all with people, car or environmental correclation.Therefore, alternatively, it is described prompting associated data can include with this threeAny one or more related data in person, for example, the vehicle status data related to car, the external rings with environmental correclationBorder data, the user data related to people.
Wherein, for vehicle status data, can include：Speed, vehicle interior temperature, pass through onboard diagnostic system (OBD, On-Board Diagnostic) or the data such as failure that obtain of data bus of vehicle (such as CAN)., can for external environment condition dataWith including：The data such as weather, visibility, condition of road surface.For user data, can include：The data such as age, sex.
Further, since current vehicle has been provided with polytype sensor in itself, therefore can be by being passed with theseThe connection of sensor rigid line, above-mentioned corresponding data are obtained by data bus of vehicle (such as CAN, LIN etc.) and network.AndAnd can also be by vehicle-mounted input equipment by user input data.
The report event formed can include：Delay broadcast, simplify report, details decomposition, volume adjustment, word speed tuneIt is whole.Prompting associated data according to being gathered can determine that one kind therein is applicable (such as be applicable and simplify report event), orIt can also determine that multiple combinations therein are applicable (such as being applicable details decomposition+volume adjustment).The prompting associated data gatheredCan be determined with the specific relation for reporting event according to the empirical value after a large amount of statistical analyses, or can also according to user fromDefinition is set to determine, but to be used as important references safely.
For example, when user data is reflected as women, external environment condition data reflection condition of road surface complexity and relatively low visibility,Combination of the report event for details decomposition+word speed adjustment+volume adjustment can be determined, that is, is decomposed and is reported content, suitably reduces languageSpeed simultaneously improves volume, and content is reported to provide clear, succinct segmentation.
In another example when vehicle status data reflection vehicle interior temperature is high, external environment condition data reflect road traffic congestion, canTo determine that report event as the combination for simplifying report+word speed adjustment, that is, is simplified and reports content and suitably reduce word speed, to releive,Compactly reported.
For another example when vehicle status data reflection is broken down, external environment condition data reflect rainstorm weather and user dataWhen being reflected as women, it may be determined that event of reporting is to simplify the combination of report+word speed adjustment+volume adjustment, that is, is simplified in reportHold, suitably reduce word speed and improve volume, to provide clear, succinct report content.
In addition, according to the difference of practical application, the action of collection prompting associated data both can be carried out independently, can also beStart again after obtaining voice message request.
The realization side of dynamic voice reminding method of the present invention is further illustrated below by way of the citing of some actual scenesFormula.
Scene --- press the voice message after the speech recognition button of in-car
Reach certain speed in vehicle, in the case where fair speed travels (such as 140km/h), the voice of speech recognitionPrompting should dynamically adjust report strategy, succinctly prompt important content.According to dynamic voice reminding method of the present invention, can detectSpeed, to determine that current report event is reported as simplifying under scene of running at high speed, and and then determine in scene of running at high speedUnder voice message content should be deleted, only retain a user and needed most in driving or most-often used content (these contentsDefinition can be based on experience value or by User Defined), omit some and drive the not high contents of relevance with current.Individually belowFor the contrast according to prior art and the voice message content of the embodiment of the present invention：
A kind of typical voice message content according to prior art is --- order please be say.Call, please say and " dialNumber " " call " in other words followed by name；Weather is inquired about, please be say " weather " or " local weather "；Stock information is inquired about, please be sayYour number to be inquired about of " stock number " heel, in other words " stock "；Peripheral facility information is inquired about, you can say that " periphery is setApply ", you also can directly say specific peripheral facility type；" news " please be said by inquiring about news information, and you also can directly say specificallyNews type；Transport information is inquired about, " traffic " please be say, to inquire about the transport information of fixed route, please be say " my route "；WillKnow more orders, please say " extra to help ".
And the voice message content after dynamic adjusts according to embodiments of the present invention is --- order please be say.Call,It please say that " dialing " " calls " followed by name in other words；The transport information of fixed route is inquired about, please be say " my route "；It is to be understood thatMore orders, it please say " extra to help ".
By contrasting the voice message content it can be found that after dynamic adjusts according to embodiments of the present invention, can to useFamily avoids listening to tediously long voice message content is divert one's attention because of concern and triggers potential safety hazard when driving to run at high speed.
Scene --- the voice message in navigation procedure
In the case of bad in the environment visual field, such as night, the voice message of navigation should dynamically adjust report strategy, succinctlyPrompt necessary information.According to dynamic voice reminding method of the present invention, the external environment condition data such as bright and dark light can be detected, to determineCurrent report event is that the details under night driving scene decompose report, and and then determines to tackle language under night driving sceneSound suggestion content and report interval (or report frequency) and be adjusted, increase reports the frequency and reduces content when reporting every time(only reporting and the maximally related content of current trip segment).Below by taking the navigation scenarios shown in Fig. 2 as an example, compare according to prior artWith the voice message content of the embodiment of the present invention：
A kind of representative voice suggestion content according to prior art is ---
It please be keep to the right [close to 300 meters of outlet], roll outlet after 300 meters away from；
Outlet please be roll away from [close to exit], then keep to the right, kept to the left after 100 meters, then turn and drive into the leftDalian Road.
And the voice message content after dynamic adjusts according to embodiments of the present invention is ---
1. please be keep to the right [close to 500 meters of outlet], outlet is rolled after 500 meters away from；
2. [close to 300 meters of outlet] drive slow please, road conditions are complicated after rolling outlet away from；
3. outlet please be roll away from [close to exit], then keep to the right；
[4. close to fork] please keep to the left now；
5. please being turned to the left [close to Dalian crossing], Dalian Road is driven into
By contrasting the voice message content it can be found that after dynamic adjusts according to embodiments of the present invention, broadcast by increaseUser's navigation hint is more efficiently given in the case of the frequency is reported so that the visual field is bad at night, content is reported by reducing, so thatObtain the content reported every time to be grasped within a very short time by user, traffic safety is also improved while avoiding diverting one's attention.
It should be noted that, although illustrate above mainly based on vehicular applications, but dynamic voice method of the present inventionBut vehicular applications are applicable not only to, it is applied equally to other electronic equipments for possessing voice recognition processing ability, such as whenThe preceding smart mobile phone for being generally integrated with speech recognition software and navigation software.For example, the side for passing through wire/wireless when smart mobile phoneAfter formula establishes communication connection with vehicle electronic device, it again may be by vehicle electronic device and its own obtains above-mentioned carryShow associated data, and according to above-mentioned processing after, pass through vehicle-mounted loudspeaker or the loudspeaker of itself carry out the report of voice message.
In addition, for vehicular applications, solution provided by the invention also includes onboard system.Shown in reference picture 3, according toThe onboard system 100 of an embodiment of the present invention, it includes：
Vehicle-mounted functional module 105, it is used to provide a user vehicular applications service；For example, vehicle-mounted functional module 105 can be withOn-vehicle information entertainment modules, with provide a user listen to such as radio station, music, on-vehicle information entertainment garment information inquiryBusiness；Or vehicle-mounted functional module 105 can also be vehicle mounted guidance module, looked into providing a user such as route planning, point of interestThe vehicle mounted guidance service such as look for；Certainly, vehicle-mounted functional module 105 can also be other conventional at present and futures will be integrated into it is vehicle-mounted shouldWith other functional modules in service；
Data acquisition module 104, its be used for gather prompting associated data, such as including：Various types of vehicles status data and/orExternal environment condition data and/or user data；It can be for example, by connecting data bus of vehicle (such as CAN, LIN etc.), connectionNetwork such as is directly connected at the collection that mode carries out above-mentioned data with various kinds of sensors rigid line；
Dynamic voice prompts detection module 103, its data gathered according to data acquisition module 104, represents to broadcast to be formedThe data of report event；
Voice message processing module 101, its by input module 107 obtain user voice message request after, according toThe data for the expression report event that dynamic voice prompting detection module 103 is formed, to determine voice message mode；
Voice message synthesis module 102, its voice message mode determined according to voice message processing module 101, synthesisExported for the voice of report, and eventually through output module 106.
In the onboard system 100 of above-described embodiment, data acquisition module 104, dynamic voice prompting detection module 103, languageSound prompting processing 101, vehicle-mounted functional module 105, voice message synthesis module 102 and input module 107, output module106, it according to having communication connection shown in Fig. 3, is passed with carrying out the data/message of each intermodule as described above between any twoIt is defeated.
In other embodiments of the invention, input module 107 and output module 106 can also be independently of above-mentioned vehicle-mountedSystem 100, such as using the contact panel of car microphone equipment or existing vehicle electronic device as external input moduleEquipment；And using vehicle-mounted loudspeaker as external output module equipment.
The function mould of part onboard system of the present invention has been integrated with view of the vehicle electronic device in current many vehiclesBlock, in actual applications can the structure based on invention onboard system to existing vehicle electronic device be designed adjustment with realityExisting function of the present invention.For example, for having possessed the onboard navigation system of speech identifying function, it is only necessary to wherein by hardware/softThe mode of part add dynamic voice prompting detection module 103, data acquisition module 104 Functional Design and for because prompting is handledThe inter-process of module 101 is adjusted, and can just form the new navigation system for possessing dynamic voice prompt facility.Certainly hereinOnly it is to make explanation become apparent from the citing made, is not limited to any application scenario of the invention and application mode.
With continued reference to shown in Fig. 4, it illustrates it and realizes that dynamic voice carries by taking the onboard system structure shown in Fig. 3 as an exampleThe process shown.
With reference to shown in reference picture 3 and Fig. 4, exemplified by method described above illustrates the scene of running at high speed of middle citing, when user is led toAfter crossing speech recognition button (input module 107) the triggering voice message request for pressing in-car, the meeting of voice message processing module 101This is handled.During this period, data acquisition module 104 can gather vehicle speed data (vehicle status data) and send to dynamicVoice message detection module 103.Dynamic voice prompts detection module 103 in analysis vehicle speed data (it is assumed that current vehicle speed isAfter 140km/h), it is determined that current report event is to run at high speed to simplify report under scene.And voice message processing module 101According to the data of the report event, it is determined that should be deleted under scene of running at high speed voice message content, omit some withIt is current to drive the not high content of relevance.And then triggering speech recognition button is obtained from sound bank (vehicle-mounted functional module 105)The content of Default sound prompting Operation afterwards, and its content is adjusted, the final above-mentioned method explanation such as that obtains refers toContent.Then, the final content, which can be sent in voice message synthesis module 102, carries out phonetic synthesis, and it can both lead toCross the software/hardware realization of text-to-speech (TTS, Text to Speech), the voice that can also be prerecorded by transferringFragment is realized.And the voice after synthesizing then carries out voice output finally by vehicle-mounted loudspeaker (output module 106), to reportTo user.
Although the present invention is disclosed as above with preferred embodiment, the present invention is not limited to this.Any art technologyPersonnel, the various changes made without departing from the spirit and scope of the present invention and modification, the protection model of the present invention all should be includedIn enclosing, therefore protection scope of the present invention should be defined by claim limited range.