CN1688126A - Multi-language speed field name and speech searching system and method based on internet - Google Patents

Abstract

This invention relates to a multi-phone clock phonetic domain name and search system and a method based on Internet, which is a technology of domain name analysis, ASR and TTS combining multi-language Internet access and the search system of Internet, phonetic CODEC, video CODEC, TCP/IP database technology and a software technology and capable of pronouncing the target network station or other platform names based on the user's voice to reach the access, capable of pronouncing the name of the searched network station and the searched key words to finish the target search. Point-to-point phonetic video communication can be realized via VOIP function in the system.

Description

Multilingual voice domain name and phonetic search system and method based on INTERNET

Technical field

The present invention relates to a kind of media system, referring to especially a kind ofly can provide multilingual voice domain name and phonetic search system and method for the user.

Background technology

At present, the application of the Internet and information technology has been deep into the every aspect of corporate behavior and personal lifestyle, many enterprises have set up enterprise application system and the website of oneself based on the Internet, be used to improve internal control, reduce cost, raise the efficiency, thereby promote Enterprises'Competitiveness; The individual also may enjoy all benefits that the Internet and information technology are brought to the full, such as widening the vision, shorten space-time distance etc.More information network and classification life website more hew out the mass advertising space of a whole page increases the performance of enterprises.Especially in recent years the development of broadband technology and popularization and application makes based on the Internet and domain name interpretation technique, speech recognition (ASR) technology, and phonetic synthesis (TTS) The Application of Technology becomes possibility.

With regard to current, that internet site is mainly finished is static, the enterprises propagandist based on literal and picture, product introduction, pre-sales support, mid-sales communication and after-sale service, need could realize by the domain name of input characters sign visiting during the user capture website.

In addition, up to the present, still lack a kind of effective user speech domain name access and search mechanisms, just can realize visit as long as allow the user say domain name, the accessed end just can utilize the content advantage and the 24-hour service advantage of internet site, for providing a kind of, the user explains based on Internet and voice domain name, the VoiceXML resolver, speech recognition engine, speech synthesis engine, voice channel, the VoiceXML gateway combines Internet, voice CODEC, video CODEC, TCP/IP, the Internet voice access search system of database technology and software engineering.Many enterprises and individuals recognize the importance that calling party is provided personalized service more and more, and Internet voice access, domain name application technology, speech recognition (ASR) technology, phonetic synthesis (TTS) technology are also more and more ripe; Voice CODEC, video CODEC and tcp/ip communication technology also develop into the practical stage.

Therefore, adapt to this application demand and technological progress just based on the multilingual voice domain name of INTERNET and phonetic search system and method and a kind of application system of proposing.

Summary of the invention

Based on the deficiencies in the prior art, main purpose of the present invention is to provide a kind of domain name interpretation technique based on Internet, speech recognition (ASR) technology, phonetic synthesis (TTS) technology, combine the multilingual internet voice visit and the search system of Internet, voice CODEC, video CODEC, TCP/IP, database technology and software engineering, this system comprises:

Control Server is used for communication signaling control and management, user management, control of authority, storage and transfer, client-side management, accounting management, data safety management;

Database server is used for data manipulation and storage;

Domain name is explained server, is used for domain name explanation and sensing;

The VoiceXML voice browser;

The VoiceXML voice gateways;

The separate customer machine is used for communication signaling control and management, Data Transmission Controlling, data safety management, language data process, video data processing, word message processing, file transfer, message (voice, literal and video) management, renewal support automatically, IVR support and hardware device support;

The assembly client, consistent with the effect of separate customer end, need in browser, move;

Internet access facility is used for the Internet and inserts;

Www server is used for information issue and collection and assembly client deployment; And

Audio and video apparatus is used for catching and changing of voice and video signal.

In the present invention, described assembly client combines with described browser address bar, explain that by VoiceXML voice browser and domain name server carries out related, when the user operates domain name and browser address bar assembly thereof, call this client by call targets name of station title, the affiliated partner of realization and domain name explanation results clauses and subclauses communicates.Described assembly client combines with described browser address bar, and by calling out or operate or clicking control element and call this client, the affiliated partner of realization and voice domain name explanation results clauses and subclauses communicates.Described VoiceXML voice browser comprises the VoiceXML resolver, speech recognition engine, speech synthesis engine and voice channel.The assembly client is embedded into and takes the form of browser component or plug-in unit element on the browser address bar.

Described system can provide one or more access modes among voice, video, literal, file, the IVR.User's input or the domain name of calling out can be other codes of the English domain name of Chinese domain name or digital domain or appointment.

The present invention also further provides a kind of multilingual voice domain name and phonetic search method based on INTERNET, comprises the steps:

System initialization;

The user calls out website (comprising search site) title log-on message system by browser address bar operation domain name or by voice channel;

Speech recognition engine obtains user speech information and transfers the computer program identifiable language to and pass through VoiceXML resolver, related domain name system;

The user is by operation domain name or assembly client, and the initiation system connects the target approach website;

By the VoiceXML browser, service end VoiceXML gateway is reported the voice webpage corresponding with the user or is obtained Search Results at user side;

Enable the VOIP function and initiate communication call;

System's end of run.

Operation domain name or assembly client, the detailed process of information system startup work is:

When other code of customer call or input Chinese domain name or digital domain or appointment or or by clicking assembly element mode invocation component client;

Assembly client and Control Server exchange of signaling;

Set up and communicate by letter with database server, www server by Control Server;

The www server is reported information and is obtained Search Results to the assembly client;

Voice access finishes.

Compared with prior art, as voice domain name and phonetic search system platform based on INTERNET, can send through visit after the title of targeted website or other website with voice according to the user, can send the title arrival search station of search website with voice, and send the searching key word voice and can finish target search.Realize dialogue with targeted sites, the VOIP function by system can also realize point-to-point voice and video communication.

In addition, the assembly client of system combines with Internet IE browser address bar, be that the assembly client is embedded into browser address bar and passes through voice domain name explanation results definite object Website page, when customer call or operation domain name and browser address bar assembly thereof, by calling out or operate or clicking control element and call this client, the affiliated partner of realization and voice domain name explanation results clauses and subclauses communicates.

The assembly client of system and domain name interpreter, the VoiceXML resolver, speech recognition engine, speech synthesis engine, the voice channel combination, promptly the assembly client is embedded into the IE browser, the assembly client possesses the RESOLVER explanation function, speech recognition (ASR) and phonetic synthesis (TTS) function are usually called this client by calling out or operate or click IE address field domain name unit, realize the affiliated partner interactive communication.Reach the voice access the Internet and finish phonetic search result's purpose.

The assembly client is embedded into the IE browser, it takes the form of control element, control element can be a button, icon, picture, literal, link, also other other visual content, the user is by direct control or call out domain name or click control element with mouse, keyboard or other input equipments and initiate related communication.

For making the present invention easier to understand, further set forth voice domain name and the phonetic search system and method that the present invention is based on INTERNET below in conjunction with accompanying drawing.

Description of drawings

Fig. 1 is the entire system block diagram;

Fig. 2 is the Control Server block diagram;

Fig. 3 is the software system structure block diagram;

Fig. 4 is the client terminal structure block diagram;

Fig. 5 is that the assembly client combines with the IE browser;

Fig. 6 shows the user in the browser form for the assembly client

Fig. 7 is that the separate customer machine combines with www server;

Fig. 8 is the workflow of assembly client when combining with the voice domain name.

Embodiment

The present invention a kind of voice domain name and phonetic search system and method based on INTERNET, it is in conjunction with the Internet technology, speech recognition (ASR) technology, phonetic synthesis (TTS) technology, the domain name interpretation technique, voice CODEC, video CODEC, TCP/IP, multiple technologies such as database technology and software engineering, for the user provides the voice access system platform, make the user from initiating voice access or phonetic search, start speech recognition engine, the voice signal that the people is spoken is converted to the Word message that can be discerned by computer program, thus identification speaker's phonetic order and word content.It is related to explain that by VoiceXML resolver and domain name server carries out, and realizes voice access or search to targeted sites, and the assembly client possesses the VOIP function, can realize point-to-point instant messaging.

Work in coordination based on the voice domain name of INTERNET and Control Server, database server, www server, name server, VoiceXML voice browser, VoiceXML voice gateways, XML, DB (the mutual modular converter of Voice Markup Language and relevant database), separate customer end and the assembly client of phonetic search system and method, complete voice access and phonetic search soft ware platform have been constituted, enterprise can be by this for the user provides intelligent service, and it is convenient and practical that the user can be experienced.Wherein, the VoiceXML voice browser comprises VoiceXML resolver, speech recognition engine, speech synthesis engine, voice channel.

1) entire system structure

Whole system is made up of Control Server, database server, www server, name server, VoiceXML voice browser, VoiceXML voice gateways, XML, DB (the mutual modular converter of Voice Markup Language and relevant database), separate customer end, assembly client computer, Internet access device, www server and audio and video apparatus.Wherein, the VoiceXML voice browser comprises VoiceXML resolver, speech recognition engine, speech synthesis engine, voice channel.

With reference to Fig. 1, in entire system structure square frame, separate customer machine 1 and VoiceXML voice browser 2 are the user application terminal, by Internet access device 10, separate customer machine 1 support voice video equipment 3, the user application terminal is interconnected or enterprise and the interconnected logical construction of enterprise's PERCOM peripheral communication of enterprises communication, and Internet access device 10 is connected with Internet.Name server 4 provides internet domain name to explain and point to service.Internet Server 5 provides Internet service.VoiceXML voice gateways 6 are finished HTML and VOICEXML conversion.Assembly client computer 2 is embedded into the IE browser and combines with it.Control Server 7 is finished the signaling control and the condition managing of communication process.Database server 8 provides the service of database sound bank.XML and DB module 9 are finished the mutual conversion of Voice Markup Language and relevant database, and the above-mentioned functions module is finished the SMF of whole system jointly.

The integral body composition of system has scalability, and separate customer end and assembly client comprise VoiceXML voice browser, database server, Control Server, Internet Server, name server, separate customer end and assembly client computer.Wherein, the VoiceXML voice browser comprises the VoiceXML resolver, speech recognition engine ASR, speech synthesis engine TTS, voice channel.VoiceXML voice gateways, XML and DB module are not quantitatively limit.

2) software system structure

Software systems adopt C/S structure, i.e. client terminal/server structure on the whole.Wherein server software moves as independently using (or service), in conjunction with software distribution formula technology, according to client terminal quantity, server software can be deployed on one or more computer, thereby balance server end flow, server is mainly finished some controlled function, and we are called Control Server to server.Client software adopts dual mode to realize: client (we are referred to as client) that promptly can independent operating and as the client (we are referred to as the assembly client) of component software form.The separate customer end is an independently application program, can isolated operation; The assembly client is an ActiveX control, need move in other Application Containers, as browser; According to concrete application form, this ActiveX control can be deployed in the IE browser and combine the domain name of existing enterprise with it, will bind again or replace with straightaway Chinese or numeral by name server, and is related with the database generation during when assembly client operation domain name through the domain name of binding.The client and server end communicates by ICP/IP protocol.

With reference to Fig. 2, in the software system structure block diagram, software systems comprise Control Server and client two parts on the whole, Control Server adopts the project organization of layering, be that Control Server is by Data Transmission Controlling B02, data safety management B03, communication signaling control and management B04, use processing layer B12, phonological component B06, video section B07, storage and transfer B08, control of authority B09, subsystems such as client-side management B10) etc. part is formed, wherein, use processing layer B12 and comprise the VoiceXML resolver, speech recognition engine ASR, speech synthesis engine TTS, voice channel.Client computer also adopts the hierarchical design structure, and has corresponding layering processing with server end.Client-server is by transport network layer (the ICP/IP protocol stack carries out direct data communication), and in addition, system also comprises database server (B11).

3) Control Server function and realization

The major function of Control Server comprises: communication signaling control and management, user management, control of authority, storage and transfer, client-side management, accounting management, data safety management etc.Communication signaling is by autonomous Design, be used to support client calls, seek, connect, State Control, exchanges data and transfer, fire compartment wall detecting and penetrate etc., the VoiceXML resolver, speech recognition engine ASR, speech synthesis engine TTS, speech recognition and conversion and control etc., the SOCKET interface function that adopts the ICP/IP protocol stack to provide is realized; User management mainly manages user's related data and data, each user (comprising enterprise customer and personal user) who uses native system has at least one independently identification number, control of authority is used to realize rights management and control, storage and in be diverted to the management ephemeral data, client-side management is realized client release control and automatic version updating function, and user management, control of authority, storage and transfer and client binding data stock technology realize; Accounting management is used for the deployment in future that system's visit quantification maybe may be used to charge, data safety management is used to support the safe transmission of data, the safe transmission of data adopts general symmetric key algorithm and asymmetric key algorithm mechanism, and specific algorithm relates to RSA, MD5 and DES/3DES.

With reference to Fig. 3, in the Control Server block diagram, the controlled function server adopts the hierarchy design.Data Transmission Controlling layer C02 finishes transfer of data by the TCP/IP development interface that transport network layer (ICP/IP protocol stack) C01 provides.The interface that data safety management layer C03 adopts Data Transmission Controlling layer C02 to provide is finished the safe transmission of data.The interface that the control of communication signaling and management C04 adopt data safety management layer C03 to provide is finished the transmission of communication signaling and data.Application function layer segment (comprising: client-side management C07, storage and transfer C06, user management 3010, control of authority C08, accounting management C09) adopts the control of communication signaling and management C04, data safety management layer C03, Data Transmission Controlling layer C02 to realize concrete function, and the application function layer segment also needs binding data storehouse system simultaneously.The separate customer machine C13 of system, assembly client computer C14, VoiceXML resolver, speech recognition engine ASR, speech synthesis engine TTS, voice channel C15 mainly communicates by application function layer and Control Server.

4) client functionality and realization

The realization of client comprises separate customer end and assembly client dual mode, dual mode is basically identical on function realizes, the separate customer end is the global function client computer, except that undertaking Resolver explanation and basic communication functions, enterprises end also can not independent operating as enterprise such as voice, video, literal, file languages identifying information compiling end, assembly client, but can be integrated in the middle of other container, as the part of other client computer.The function of client comprises: communication signaling control and management, Data Transmission Controlling, data safety management, language data process, video data processing, word message processing, file transfer, message (voice, literal and video) management, renewal support automatically, IVR support and hardware device support etc.Control of client communication signaling and the control of management and service device end communication signaling and management function cooperate, finish speech recognition ASR, speech conversion TTS, finish VOIP calling, addressing, connect, State Control, exchanges data and transfer, fire compartment wall detecting and function such as penetrate, Data Transmission Controlling is used to finish real-time Transmission control, the adaptive bandwidth control of data, the transmission of non-real-time data, and control of client communication signaling and Data Transmission Controlling adopt the realization of ICP/IP protocol stack; Client data safety management and the safety management of server end data are worked in coordination, finish the data security transmission, adopt general symmetric key algorithm and asymmetric key algorithm mechanism, asymmetric key algorithm adopts RSA, symmetric key algorithm adopts DES/3DES, and hash algorithm adopts MD5; Language data process comprises the catching in real time of voice, coding and decoding (CODEC), transmission, playback, video data is handled and to be comprised the catching in real time of video, coding and decoding (CODEC), transmission, play, and the coding of voice and video is followed the voice and video coding and decoding algorithm of the world; The Word message transmission that word message is finished dealing with instant, file transfer is used for the transfer files data, and word message is handled and file transfer all adopts the mode of ciphertext to transmit; Automatically upgrade the auto-update of supporting to finish client software; IVR supports to finish interactively voice, video, picture and text response, and an end that communicates to connect is according to the operation of the other end, from dynamic response, plays or shows voice, video, graph-text content to the other end.

With reference to Fig. 4, in the client terminal structure block diagram, the design of customer end adopted hierarchy, data transfer layer D02, data safety management D03, communication signaling control and management D04 and Control Server are corresponding substantially, but in client data transport layer D02, realize real-time Transmission control, Adaptive Transmission control and non real-time transmission control.At application function layer D05, comprise version updating, word message, file transfer, message management, IVR management; Speech recognition ASR, the processing capacity of speech conversion TTS and Voice ﹠ Video.

5) the assembly client combines with the IE browser

The assembly client can be browsed the operation of container environment at IE, can pass through the client browser script, realizes combining with the IE browser as the client of component software form, directly is created on the IE frame, and the assembly client is activated when IE starts.

With reference to Fig. 5, browse in the container combination in assembly client and IE, Internet user utilizes browser (E01) user to call out domain name or operate domain name at address field at voice channel, explain that through VoiceXML resolver or Resolver DNS realizes the voice access WWW of enterprise website or obtains target search, related Control Server of assembly client and database server, Control Server converts database server to VXML to HTML about the information of website by the VoiceXML voice gateways, and client is obtained voice messaging.The assembly client is finished the visit of targeted sites or phonetic search.

6) client independent utility

The separate customer end can use separately, explain VoiceXML resolver and the basic communication functions except that undertaking Resolver, also as enterprise's languages identifying information compiling ends such as voice, video, literal, files. the terminal that the enterprise site of mainly providing and delivering accessed by the user uses, can be used for compiling voice and video, making needs the file at first play when accessed, sets the management of accessed file or VoiceXML voice gateways.

Above disclosed only for the present invention is based on the voice domain name of INTERNET and the preferred embodiment of phonetic search system and method, certainly can not limit the present invention's interest field with this, therefore the equivalent variations of being done according to the present patent application claim still belongs to the scope that the present invention is contained.

Claims (9)

1. multilingual voice domain name and phonetic search system based on an INTERNET, it is characterized in that: this system comprises:

Control Server is used for communication signaling control and management, user management, control of authority, storage and transfer, client-side management, accounting management, data safety management;

Database server is used for data manipulation and storage;

Domain name is explained server, is used for domain name explanation and sensing;

The VoiceXML voice browser;

The VoiceXML voice gateways;

The separate customer machine is used for communication signaling control and management, Data Transmission Controlling, data safety management, language data process, video data processing, word message processing, file transfer, message (voice, literal and video) management, renewal support automatically, IVR support and hardware device support;

The assembly client, consistent with the effect of separate customer end, need in browser, move;

Internet access facility is used for the Internet and inserts;

Www server is used for information issue and collection and assembly client deployment; And

Audio and video apparatus is used for catching and changing of voice and video signal.

2. multilingual voice domain name and phonetic search system based on INTERNET according to claim 1, it is characterized in that: described assembly client combines with described browser address bar, explain that by VoiceXML voice browser and domain name server carries out related, when the user operates domain name and browser address bar assembly thereof, call this client by call targets name of station title, the affiliated partner of realization and domain name explanation results clauses and subclauses communicates.

3. multilingual voice domain name and phonetic search system based on INTERNET according to claim 2, it is characterized in that: described assembly client combines with described browser address bar, by calling out or operate or clicking control element and call this client, the affiliated partner of realization and voice domain name explanation results clauses and subclauses communicates.

4. multilingual voice domain name and phonetic search system based on INTERNET according to claim 3, it is characterized in that: described VoiceXML voice browser comprises the VoiceXML resolver, speech recognition engine, speech synthesis engine and voice channel.

5. according to claim 3 or 4 described multilingual voice domain name and phonetic search systems based on INTERNET, it is characterized in that: the assembly client is embedded into and takes the form of browser component or plug-in unit element on the browser address bar.

6. according to claim 1 to 4 each described multilingual voice domain name and phonetic search system based on INTERNET, it is characterized in that: described system can provide one or more access modes among voice, video, literal, file, the IVR.

7. according to claim 1 to 4 each described multilingual voice domain name and phonetic search system based on INTERNET, it is characterized in that: user's input or the domain name of calling out can be other codes of the English domain name of Chinese domain name or digital domain or appointment.

8. multilingual voice domain name and phonetic search method based on an INTERNET is characterized in that comprising the steps:

System initialization;

The user calls out website (comprising search site) title log-on message system by browser address bar operation domain name or by voice channel;

Speech recognition engine obtains user speech information and transfers the computer program identifiable language to and pass through VoiceXML resolver, related domain name system;

The user is by operation domain name or assembly client, and the initiation system connects the target approach website;

By the VoiceXML browser, service end VoiceXML gateway is reported the voice webpage corresponding with the user or is obtained Search Results at user side;

Enable the VOIP function and initiate communication call;

System's end of run.

9. a kind of multilingual voice domain name and phonetic search method based on INTERNET according to claim 8 is characterized in that: operation domain name or assembly client, and the detailed process of information system startup work is:

When other code of customer call or input Chinese domain name or digital domain or appointment or or by clicking assembly element mode invocation component client;

Assembly client and Control Server exchange of signaling;

Set up and communicate by letter with database server, www server by Control Server;

The www server is reported information and is obtained Search Results to the assembly client;

Voice access finishes.

Images