CN1256835C - A system and method for realizing voice Internet access - Google Patents

Abstract

The present invention discloses a phonetic Internet accessing system which comprises a call accessing unit which is connected by a local area network, a call distribution unit, a call processing unit and an executing platform, wherein the call processing unit is connected with an Internet content supply platform through the Internet and further comprises an explanation module of a phonetic language process file, wherein the explanation module is used for obtaining and explaining a phonetic language process file. The present invention simultaneously discloses a method for realizing phonetic Internet access. When a user dials accessing numbers and makes a call, the system receives and distributes the call of the user to the designated call processing unit; the call processing unit obtains and explains the phonetic language process file according to the call of the user, and plays an explanation result to the user. The system and the method provide a popular, economic and convenient accessing mode for the user, and bring great economic benefit for telecommunication operators, Internet operators and Internet content providers.

Description

A system and method for realizing voice Internet access

technical field

The invention relates to an Internet access technology, in particular to a system and method for realizing voice Internet access.

Background technique

In recent years, the Internet has gradually become an indispensable tool in people's daily life due to its characteristics of real-time, ease of use, and rich content. People can send and receive emails, obtain news in real time, and inquire about weather, flights, stocks, etc. information. The emergence of the Internet shortens the distance between people, enhances the communication between people, and enriches people's lives, so it has developed rapidly. However, at present, the Internet is accessed through text input through computers or mobile terminals, which requires users to have hardware devices such as computers or mobile terminals with Internet access functions and corresponding Internet browsing software, and requires users to have relatively professional Internet access skills. Undoubtedly, many people are blocked from the Internet, which is a defect for users and Internet operators. Therefore, people are in urgent need of a more convenient and economical way of surfing the Internet to better meet the needs of more people to obtain information quickly and in large quantities in the information society.

The telephone intelligent customer service center, which provides services for the largest number of telephone users, has also developed rapidly. Through the intelligent customer service center, users can inquire about information such as stocks, weather, and flights, which brings great convenience to their lives. Figure 1 is a schematic diagram of the network configuration structure of the current intelligent customer service center. It can be seen from the figure that the current intelligent customer service center includes a call access queuing device (ACD) and a communication front-end server (CTI Server) as a call access unit. ), the customer center server (CCS) as a call distribution unit, the automatic voice response system (IVR) as a call processing unit, and the voice processing station (VP) as an execution platform, all of which are connected to the local area network. The switched telephone network/public land mobile network (PSTN/PLMN) is connected to the user's fixed telephone or mobile terminal. When the user dials the access number of the customer service system, the user's call is connected to the call control CCS through the ACD and CTI Server, and the CCS distributes the call to a certain IVR. Execute corresponding business processes, and control VPs and users to perform interactive operations such as playing announcements and receiving numbers.

IBM and other companies proposed the Voice Extensible Markup Language (VoiceXML) in early 1999, which is a transmission standard for accessing Internet web page content and its interactive voice response through voice dialogue, which simplifies the personality of voice response services on the web page The creation of an intelligent interface enables people to access information and services on the website through voice and telephone, and at the same time, it is combined with the script of the Common Gateway Interface (CGI) to retrieve the back-end database and access the intranet of the enterprise, so that the voice browser is finally combined with Traditional language browsers are integrated to realize the perfect combination of computer network and telephony technology.

Since people long for more popular, more economical and more convenient access to the Internet, connecting the currently most widely used telephone network with the Internet through Voice Extensible Markup Language will become the best way to solve this demand.

Contents of the invention

The main purpose of the present invention is to provide a voice Internet access system, which can realize voice Internet access more popularly, more conveniently and more economically, and provide users with more convenient services.

Another object of the present invention is to provide a method for implementing voice access to the Internet, so that users can directly obtain required information from the Internet through voice, and the operation is simple and easy to implement.

The purpose of the present invention is achieved through the following technical solutions:

A voice Internet access system, including a call access unit connected to a local area network, a call distribution unit, a call processing unit and an execution platform, the call access unit is connected to the telephone network for accessing user calls, and the call distribution unit is used for The incoming user call is distributed to the call processing unit, which is used to process the user's call and transmit the processing result to the execution platform, and the execution platform is used to execute the processing result and interact with the user, wherein the system further includes a connection to the local area network The text-to-speech conversion unit is used to convert the text from the call processing unit into voice and play it to the user through the execution platform. Converting to voice; the call processing unit is also connected to the Internet content providing platform through the Internet, and further includes an interpretation module of the voice language flow file for obtaining and interpreting the voice language flow file.

In the above-mentioned voice Internet access system, it may further include a voice recognition unit connected to the local area network, used to identify the voice input by the user collected by the execution platform, and send the voice recognition result to the call processing unit, the execution platform further includes a voice recognition module, It is used to complete automatic speech recognition with the cooperation of an independently set speech recognition unit.

In the above-mentioned voice Internet access system, the voice language process file used is a voice extensible markup language process file.

A method for realizing Internet access by voice, at least including the following steps:

a. After the system receives the voice Internet call initiated by the user terminal, it assigns the user call to the designated call processing unit;

b. The call processing unit obtains the voice language flow file from the Internet content providing platform according to the user call and interprets the voice language flow file;

c. judge whether the content specified to be played in the speech language process file after the explanation is a text, if it is a text, then perform text-to-speech conversion on the text, and then play the interpreted result of the text-to-speech conversion to the user, otherwise directly to the user User play interprets the results.

In the above-mentioned method for realizing Internet access by voice, the following steps are further included: after the system plays the explanation result to the user, it continues to collect user input, and judges whether the content requested by the user is included in the voice language obtained from the Internet content providing platform according to the user input. In the flow file, if the content requested by the user is included in the speech language flow file, continue to interpret the speech language flow file according to the user request; if the content requested by the user is not included in the speech language flow file, from the Internet content Provide a platform to reacquire the corresponding speech language process files and explain them.

In the above-mentioned method for realizing Internet access by voice, the following steps are further included: the system judges whether the collected user input is voice, if the user input is voice, perform voice recognition on the voice, and then obtain and explain the voice language process file, otherwise directly obtain And interpret the speech language flow file.

It can be seen from the above technical solutions that the terminal used to access the Internet is the telephone, and the line is the telephone network, and the telephone and the telephone network have lower cost and higher popularity than computers and mobile terminals with Internet access functions. Rate. For users, they can access the content of the Internet through keystrokes or voice, and can be used very easily without special skills, so it is more convenient to use. At the same time, users can send and receive e-mails and consult Internet information anytime and anywhere through ordinary telephones, mobile terminals, car phones and other communication devices, breaking through the limitations of time and space, and truly realizing "easy access to the Internet". Therefore, for users, it is the most convenient and economical to utilize the present invention to surf the Internet by telephone.

After using the system and method of the present invention, any user with a telephone can realize surfing the Internet through telephone voice. Since the PSTN/PLMN network has been fully developed, their combination with the Internet network will inevitably provide them with a very broad business space and bring them huge profits for telephone system operators and Internet operators . Similarly, for Internet content providers, they can attract more people to visit their webpages, and they can bring greater profits to themselves by means of charging content or increasing advertising charges. Therefore, for equipment providers, telecommunication and Internet operators, and Internet content providers, utilizing the present invention to surf the Internet through the telephone will bring them huge economic benefits.

Description of drawings

Fig. 1 is a schematic diagram of a network configuration structure of an intelligent customer service center according to the prior art;

Fig. 2 is a schematic diagram of the network configuration structure of the telephone voice Internet access system according to the present invention;

Fig. 3 is the flow chart of the method for realizing Internet access by telephone voice according to the present invention;

Fig. 4 is a diagram of the call processing process between hardwares of the method for realizing Internet access by telephone voice according to the present invention.

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

Fig. 2 is a schematic diagram of the network configuration structure of the telephone voice Internet access system according to the present invention. It can be seen from the figure that in addition to the ACD, CCS, CTI Server, VP and other units connected to the local area network in the existing intelligent customer service center, it also includes the voice information engine server (VoiceIE), text language Conversion server (TTS Server), automatic speech recognition server (ASRServer) and Internet content provider (ICP), wherein VoiceIE is also connected with the Internet, and accesses the web server of ICP through the Internet. In this system, ACD and CTI Server are the call access unit, CCS is the call distribution unit, VoiceIE is the call processing unit, VP is the execution platform, the text-to-speech server is the text-to-speech unit, and the automatic speech recognition server is the speech recognition unit. unit.

Among these newly added units, TTS Server is used to cooperate with VP to complete the text-to-speech conversion function, and ASR Server is used to cooperate with VP to complete the automatic speech recognition function. In the present invention, in addition to the original functions of sound playback, number collection, and recording, VP adds functional modules such as speech recognition and text-to-speech conversion. The ICP is used to store the published Speech Extensible Markup Language process, and provide the corresponding Speech Extensible Markup Language process according to the request of the voice information engine server, and complete the requirements in the Speech Extensible Markup Language process through the Common Gateway Interface (CGI) program logic function.

In the system according to the present invention, VoiceIE is the core unit, which is used to explain and execute the business logic process distributed in the intelligent customer service center, wherein the business logic supports the business-independent component (SIB) process and the voice extensible markup language process two ways. In order to realize surfing the Internet through voice, the voice extensible markup language flow file is used in the present invention. For the Voice Extensible Markup Language process file, VoiceIE obtains the corresponding Voice Extensible Markup Language process file from the Internet content providing platform, interprets and executes the Voice Extensible Markup Language process, and cooperates with VP to complete the collection of user input and provide the user with corresponding Output information and other interactive processes.

Fig. 3 is a flow chart of a method for realizing Internet access by telephone voice according to the present invention. It can be seen from the figure that when the user uses the above-mentioned voice Internet access system to access the Internet through voice, he first dials the access number of the system, and then performs the following steps:

Step 301, the user's call is connected to the CCS controlled by the call through the ACD and the CTI Server.

In step 302, the CCS distributes the user's call to a certain VoiceIE.

Step 303, VoiceIE determines the Uniform Resource Locator (URL) corresponding to the Voice Extensible Markup Language process according to the preset configuration file, and then requests the ICP webpage server corresponding to the URL to provide the content required by the user.

Step 304, VoiceIE obtains from the ICP webpage server the voice Extensible Markup Language flow file provided corresponding to the user's request.

Step 305, VoiceIE interprets the voice extensible markup language flow, and sends the interpretation result to the VP.

In steps 306-308, the VP judges whether the provided content is text according to the voice content requested by the VoiceIE. If the content provided is text, then the VP transmits the text content to the text-to-speech conversion server, and the text-to-speech conversion server converts the text into speech, and returns the speech after the conversion to the VP, and then enters step 309; if the provided content If it is not text, that is, it is voice directly, step 309 is performed directly.

In step 309, the VP plays the voice to the user.

If the user still needs to continue to obtain information, the user will continue to interact with the system, and then continue to execute step 310 .

Step 310, VP collects user input.

In steps 311-313, the VP judges whether the user input is voice. If it is voice, the VP collects the voice data of the user in real time, and transmits it to the voice recognition server, and then the voice recognition server recognizes the voice, and reports the recognition result to the VP, and enters step 314; otherwise, directly enters step 314.

Step 314, VP reports the recognition result to VoiceIE.

Step 315, VoiceIE determines whether the content requested by the user is included in the received VoiceXML file according to the user input. If the content requested by the user has been included in the voice extensible markup language flow file received in step 304, then execute step 305 and subsequent steps, otherwise execute step 303 and subsequent steps until the user hangs up or no longer inputs new instructions.

Fig. 4 is a diagram of the call processing process between hardwares of the method for realizing Internet access by telephone voice according to the present invention. The following describes the method of using the above-mentioned system to realize Internet access through voice in combination with FIG. 3 and FIG. 4 and specific embodiments.

Specific embodiment 1:

Take the user querying weather conditions as an example, set the file that VoiceIE obtains from the ICP web server as text voice, and the user uses voice input to interact with the system.

If the user wishes to inquire about weather conditions through the voice Internet access system, first dial the system access number, such as 100, and in step 301, the user's call is connected to the CCS controlled by the call through the ACD and CTI Server. In step 302, the CCS distributes the user's call to a certain VoiceIE. In step 303 , VoiceIE determines the Uniform Resource Locator (URL) corresponding to the voice extensible markup language process according to the preset configuration file. The ICP webpage server corresponding to the URL is requested to provide the content required by the user. In step 304, VoiceIE obtains the voice extensible markup language flow file corresponding to the user request from the ICP web server. In step 305, VoiceIE interprets the Voice Extensible Markup Language flow, and sends the interpretation result to the VP. For example, the voice XML flow file contains the content of playing the text voice "Welcome to the Weather Forecast Center, what do you want to know?"Where's the weather?" In step 306, the VP judges that the provided content is text according to the content of the voice played by the VoiceIE request. In step 307, the VP transmits the text content to the text-to-speech server. In step 308, the text-to-speech server converts the text to speech and returns the converted speech to the VP. In step 309, the VP plays the voice to the user, then the user will hear "Welcome to the weather forecast center, where do you want to know the weather?" 310. The VP collects voice input from the user. In step 311, the VP judges that the user's input is voice, and in step 312, collects the user's voice data in real time and transmits it to the voice recognition server. In step 313, the speech recognition server recognizes the user's speech and reports the recognition result to the VP. In step 314, the VP reports the recognition result to VoiceIE, and VoiceIE knows that the user wishes to inquire about the weather conditions in Shenzhen. Now VoiceIE judges whether the weather condition of Shenzhen is included in the speech extensible markup language flow file that receives from ICP webpage server in the front step 303, in this embodiment, the weather condition of Shenzhen has been included in the speech extensible markup that receives in front language flow file, the system re-enters step 305 and starts the execution of subsequent steps. For example, in step 305, VoiceIE interprets the speech XML process and sends the interpretation result to the VP. For example, the voice XML process file contains the text "Excuse me, what time do you want to know the weather conditions in Shenzhen? "In step 306, VP judges that the provided content is text according to the voice content of the VoiceIE request to play. In step 307, the VP transmits the text content to the text-to-speech server. In step 308, the text-to-speech server converts the text into speech, and returns the converted speech to the VP. In step 309, the VP plays the voice to the user, then the user will hear "May I ask what time is the weather in Shenzhen?" At this time, the user can tell the system "tomorrow" by voice, and in step 310, the VP collects the user's Voice input. In step 311, the VP judges that the user input is voice, then in step 312, collects the user's voice data in real time, and transmits it to the voice recognition server. In step 313, the speech recognition server recognizes the user's speech and reports the recognition result to the VP. In step 314, the VP reports the recognition result to VoiceIE, so that VoiceIE knows that the user wishes to inquire about the weather conditions in Shenzhen the next day. Then in step 315, the system then judges whether the weather condition of the next day in Shenzhen is included in the speech extensible markup language process file, for example, it is not included in the speech extensible markup language process file received in front, then the system continues to execute the steps 303. VoiceIE continues to determine the Uniform Resource Locator (URL) corresponding to the Voice Extensible Markup Language process according to the preset configuration file, and then requests the ICP webpage server corresponding to the URL to provide the content required by the user. In step 304, VoiceIE obtains the voice extensible markup language flow file corresponding to the user request from the ICP web server. Then in step 305, VoiceIE interprets the voice extensible markup language flow, and sends its interpretation result to the VP. Then, the system continues to execute step 306 and subsequent steps until the user hangs up or does not input new instructions.

Specific embodiment 2:

Taking the user querying sports news as an example, the file obtained by VoiceIE from the ICP web server is set to be a voice file, and the user uses voice input to interact with the system.

If the user wishes to query sports news online by voice, he can directly dial the system access number 100, and at this moment in step 301, the user's call is accessed to the CCS controlled by the call through the ACD and CTI Server. In step 302, the CCS distributes the user's call to a certain VoiceIE. In step 303, VoiceIE determines the Uniform Resource Locator (URL) corresponding to the voice extensible markup language process according to the preset configuration file, if the webpage of the ICP corresponding to the URL has multiple contents, such as including weather, sports and stock quotations and other columns, the system will ask the user to choose. In step 304, VoiceIE obtains from the ICP webpage server the voice Extensible Markup Language flow file provided by the ICP and required to be selected by the user. In step 305, VoiceIE interprets the Voice-XML process and sends the interpretation result to the VP, for example, the voice-XML process file contains the voice "Please select the content you care about, weather, sports, stock quotes. ....." In step 306, it is judged that the audio extensible markup language process file contains audio, so it can be directly played to the user through the VP, and text-to-speech conversion is no longer required, thereby omitting steps 307 and 308. In step 309, the VP directly plays the voice to the user "Please select the content you care about, weather, sports, stock market..." At this time, the user can tell the system through voice "I want to know the current Men's Football League One point table", in step 310, the VP collects the user's voice input. In step 311, the VP judges that the user input is voice, and in step 312, the VP collects the user's voice data in real time and transmits it to the voice recognition server. In step 313, the speech recognition server recognizes the user's speech and reports the recognition result to the VP. In step 314, the VP reports the recognition result to VoiceIE, so that VoiceIE knows that the user wants to inquire about sports and the content of the men's football League One score table. Then, in step 315, VoiceIE judges whether the content requested by the user is included in the voice extensible markup language flow file received in step 304, and continues to execute step 303 and subsequent steps or step 305 and subsequent steps according to the judgment result until the user Hang up or no new commands are input.

Specific embodiment 3:

Taking the user querying sports news as an example, the file obtained by VoiceIE from the ICP web server is set to be a voice file, and the user uses key input to interact with the system. At the same time, users can jump between different ICPs.

In the example where the above-mentioned user wishes to query sports news online by voice, the user directly dials the access number 100, and at step 301, the user's call is accessed to the CCS controlled by the call through the ACD and CTI Server. In step 302, the CCS distributes the user's call to a certain VoiceIE. In step 303, VoiceIE determines the Uniform Resource Locator (URL) corresponding to the voice extensible markup language process according to the preset configuration file, if the webpage of the ICP corresponding to the URL has multiple contents, such as including weather, sports and stock quotations and other columns, the system will ask the user to choose. In step 304, VoiceIE obtains from the ICP webpage server the voice Extensible Markup Language flow file provided by the ICP and required to be selected by the user. In step 305, VoiceIE interprets the voice extensible markup language process, and sends the interpretation result to the VP, for example, the content of the voice extensible markup language process file is voice "Please select the content you care about, press 1 for weather, press 2 For sports, press 3 for stock quotes..." In step 309, the VP directly plays a voice to the user "Please select the content you care about, press 1 for weather, press 2 for sports, press 3 for stock quotes... ...." At this time, the user selects by pressing the key, for example, pressing 2 to tell the system "sports". At step 310, the VP collects user input. In step 311, it is judged that the user input is a key input, so voice recognition is not required, thus steps 312 and 313 are omitted. In step 314, the VP directly reports the recognition result to VoiceIE, so that VoiceIE knows that the user wishes to inquire about sports content. Then, continue to execute step 315 and continue to execute subsequent steps according to the judgment result of step 315. After several interactions, the user can get the sports content he wants to know. If the user wants to know other ICP content, he can tell the system that he wants to know the sports news of another ICP by voice, or select another ICP under the system prompt by key input. After the execution platform sends the user input to VoiceIE, VoiceIE will The user needs to obtain the speech extensible markup language process file from another ICP web server designated by the user, and continue to execute step 303 and subsequent steps until the user hangs up or no new instructions are input.

According to the technical solution of the present invention, it can be known that the functions of each server in the present invention can be integrated to a certain extent, for example, the functions of the speech recognition server and the text-to-speech conversion server can be placed in one server. Moreover, the system of the present invention may include fax processing desks and seats to provide more perfect and thoughtful services to users. Therefore, it can be understood that the specific embodiments according to the present invention are only exemplary, and are not intended to limit the protection scope of the present invention.

Claims (6)

1. A voice Internet access system, comprising a call access unit connected to the local area network, a call distribution unit, a call processing unit and an execution platform, the call access unit is connected to the telephone network, and is used to access the user's call, and the call distribution unit is used For distributing the incoming user calls to the call processing unit, the call processing unit is used to process the user's call and transmit the processing result to the execution platform, and the execution platform is used to execute the processing result and interact with the user. It is characterized in that the system It further includes a text-to-speech conversion unit connected to the local area network, which is used to convert the text from the call processing unit into voice and play it to the user through the execution platform. The text-to-speech conversion is completed with the cooperation of the system; the call processing unit is also connected to the Internet content providing platform through the Internet, and further includes an interpretation module of the speech language flow file, which is used to obtain and interpret the speech language flow file. 2. The voice online system according to claim 1, characterized in that, the system further includes a voice recognition unit connected to the local area network, used to identify the user input voice collected by the execution platform, and send the voice recognition result to the call processing unit, the execution platform further includes a speech recognition module, which is used to complete automatic speech recognition with the cooperation of an independently set speech recognition unit. 3. The voice surfing system according to claim 1, characterized in that, the voice language process file is a voice Extensible Markup Language process file. 4. A method for realizing voice surfing the Internet is characterized in that it at least comprises the following steps: a. After the system receives the voice Internet call initiated by the user terminal, it assigns the user call to the designated call processing unit; b. The call processing unit obtains the voice language flow file from the Internet content providing platform according to the user call and interprets the voice language flow file; c. judge whether the content specified to be played in the speech language process file after the explanation is a text, if it is a text, then perform text-to-speech conversion on the text, and then play the interpreted result of the text-to-speech conversion to the user, otherwise directly to the user User play interprets the results. 5. The method according to claim 4, further comprising the step of: after the system plays the explanation result to the user, it continues to collect user input, and judges whether the content requested by the user is included in the content obtained from the Internet content providing platform according to the user input. In the speech language flow file, if the content requested by the user is included in the speech language flow file, continue to interpret the speech language flow file according to the user request; if the content requested by the user is not included in the speech language flow file, The corresponding speech language process file is reacquired from the Internet content providing platform and explained. 6. The method according to claim 5, further comprising the steps of: the system judges whether the collected user input is voice, and if the user input is voice, performs voice recognition on the voice, and then obtains and interprets the voice language flow file, otherwise directly acquire and interpret the speech language flow file.

Images