JPH08331618A - Private branch telephone system and information processing method for same - Google Patents

Abstract

PURPOSE: To improve convenience for a user by automatically transferring a telephone call while being linked with a computer by identifying a speaker corresponding to the voice of the user and generating a word group for guidance, etc., decided in advance for the user. CONSTITUTION: An application program on automatic transfer is executed. Then, the screen of a display device at an extension terminal (multi-functional telephone set) 6 of the user (caller or callee) is turned on. The user transmits an instruction on the distribution on ordinary time and absent time to a computer 7 according to the instruction on the screen or the guidance of synthetic voice from a voice synthesizer 12. The computer 7 preserves those instructed contents in an external storage device 8. Next, it is discriminated whether the input is completed or not and in the case of end, call incoming to the user is waited. When there is an incoming call, the guidance is let flow from the voice synthesizer 12. When the caller originates a call, speaker identifying processing is performed. When any relevant person is found in registered persons by the identifying processing, it is judged whether the transfer is instructed to the relevant person or not and when it is instructed, the processing of transfer is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a private telephone system and a private telephone which provide various services by connecting an automatic private branch exchange (hereinafter referred to as "PBX") and a computer and controlling the PBX from the computer. The present invention relates to a system information processing method.

[0002]

2. Description of the Related Art FIG. 13 is a block diagram showing a conventional private telephone system. In FIG. 13, reference numeral 1 denotes a PBX which is installed in the premises and exchanges and connects an external telephone (station telephone) with an extension telephone and extension telephones. Reference numeral 2 denotes a station line (external line) which connects the PBX 1 and the public telephone network 3. Is an extension that connects the PBX 1 to the extension telephone 5 and the multifunction telephone 6.

The function of the private telephone system configured as described above will be described. The user of the extension 4 who is on the premises can make a call between the extensions through the PBX 1.
The extension telephone 5 connects to the central office line 2 to access the public telephone network 3 to make a call with an external telephone.

[0004]

However, in the above-mentioned conventional private telephone system, the extension telephone 5 and the multi-function telephone 6 can only provide the function of making or receiving a call as described above. There are also those that provide functions such as receiving messages while the user is absent, such as an answering machine, but these functions are, of course, much more natural than human processing functions. There was a problem that it was limited.

The present invention solves the above-mentioned conventional problems. By incorporating a voice recognition device and a voice synthesis device into a computer, the telephone agency work normally performed by a human secretary is automatically performed to a certain extent. Information processing of a private telephone system that can automatically perform to some extent a telephone intermediary work normally performed by a human secretary by incorporating a voice recognition device and a voice synthesizer into a computer. The purpose is to provide a method.

[0006]

To achieve this object, a private branch telephone system according to claim 1 of the present invention is connected to an automatic private branch exchange and an automatic private branch exchange through a dedicated interface, and a voice input / output interface is provided. A computer that has a voice recognition device that is connected to the computer through a voice input / output interface and that identifies the speaker by the voice of the user, and a computer that is connected through the voice input / output interface for predetermined guidance, etc. A speech synthesizer that generates a group of words for a user, a storage device that has a database that contains a group of rules related to telephone calls, and instructions for storing instructions and instructions given by the user to the agent for incoming calls. If there is no automatic intermediary means to automatically perform the intermediary rules of Is configured to automatically call an extension telephone number registered in advance from the automatic private branch exchange, search for the other party by synthetic voice, and have an automatic transfer means for transferring an incoming call to the found other party from the automatic private branch exchange. have.

According to another aspect of the present invention, a private telephone system is connected to an automatic private branch exchange, an automatic private branch exchange through a dedicated interface, a computer having a voice input / output interface, and a computer and a voice input / output interface. A voice recognition device that identifies a speaker by the voice of the user, a voice synthesis device that is connected to a computer via a voice input / output interface, and generates a predetermined word group for guidance or the like to the user, A storage device having a database accommodating a group of rules relating to telephone transfer, speaker identification by the voice of the caller, word recognition of the call destination made by the caller, and pre-registered number corresponding to the callee Searched, made a call with the searched number, and registered if there was a response to the calling of the searched number It has a configuration which includes an automatic originating agency means to request the agency telling the party name of the caller and retrieved number by adult speech.

According to another aspect of the present invention, a private branch telephone system is connected to an automatic private branch exchange, an automatic private branch exchange through a dedicated interface, a computer having a voice input / output interface, and a computer and a voice input / output interface. A voice recognition device that identifies a speaker by the voice of the user, a voice synthesis device that is connected to a computer via a voice input / output interface, and generates a predetermined word group for guidance or the like to the user, A storage device having a database accommodating a group of rules relating to telephone intermediary, and in the case of a caller whose voice is not registered in advance, the caller's utterance data is automatically recorded,
It has a configuration including an automatic registration unit that adds a sender as a registrant, and an intermediation processing unit that performs an intermediation process based on registration data of the automatic registration unit.

According to a fourth aspect of the present invention, there is provided an information processing method for a private telephone system, which comprises an automatic intermediary step for automatically performing an intermediary for an incoming telephone call according to an instruction given by a user or an intermediary rule for storing a storage device. When the other party is absent, the automatic private branch exchange automatically calls the extension telephone number registered in advance, searches for the other party by synthetic voice, and transfers the incoming call from the automatic private branch exchange to the found party. And a transfer process.

According to another aspect of the present invention, there is provided an information processing method for a private telephone system in which a speaker is identified by a voice of a caller, a callee's voice is recognized as a callee, and a preregistered number corresponding to the callee is used. Automatically request the agent by calling the searched number, and if there is a response to the searched number, the name of the sender and the partner of the searched number are announced by the registered synthetic voice. It has a configuration having a transmission intermediary process.

In the information processing method of the private telephone system according to claim 6, in the case of a caller whose voice is not registered in advance, the utterance data of the caller is automatically recorded and the caller is automatically added as a registrant. It has a configuration including a registration process and an intermediary process process for performing an intermediary process based on registered data.

[0012]

With this configuration, the speaker is identified by the voice of the user, and a predetermined word group for guidance or the like is generated for the user, so that the telephone intermediary that a human secretary performs is performed. The work of can be automatically performed, and the convenience for the user can be further enhanced.

[0013]

【Example】

(First Embodiment) A private telephone system and an information processing method of the private telephone system according to a first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a private telephone system according to an embodiment of the present invention. In FIG. 1, 1 is PB
X, 2 are office lines, 3 is a public telephone network, 4 is an extension line, 5 is an extension telephone set, and 6 is a multi-function telephone set. Since these are the same as those in FIG. . 7 is a PBX 1 through a transmission line 9 by a serial interface
Connected to the computer, 8 is an external storage device for storing an interface program or the like for controlling the PBX 1, and 10 is a multifunction telephone 6 with a serial interface.
A computer connected to the computer; 11 is a voice recognition device for identifying a speaker by the voice of the caller or the receiver; 12 is a voice for generating a predetermined word group for guidance to the caller or the receiver. It is a synthesizer.

Next, functions and connections of such a private telephone system will be described. The PBX 1 has functions such as control of outgoing and incoming calls, switching of line connection, etc., and is connected to the public telephone network 3 by a central line 2. Further, the computer 7 is connected to the PBX 1 via a transmission line 9 via a dedicated interface for connecting the PBX 1 and the computer 7. The computer 7 controls the function of the PBX 1. Further, an external storage device 8 is connected to the computer 7. As computers, there are a computer 7 connected to the PBX 1 and a computer 10 connected to the multifunction telephone 6. Further, a voice recognition device 11 and a voice synthesis device 12 are connected to the computer 7 via a built-in voice input / output interface.

FIG. 2 is a functional block diagram showing the function realizing means of the computer 7 of FIG. In FIG. 2, 71
Is an automatic intermediary means for automatically performing an intermediary for an incoming call according to an instruction given by a user or a prescribed intermediary rule stored in the external storage device 72; and 72 is an intermediary processing means in an automatic intermediary means 71 for performing an intermediary process. , 73 automatically calls the extension telephone number registered in advance from the PBX 1 when the other party who is the automatic agent is absent, searches the other party by the synthesized voice, and transfers the incoming call from the PBX 1 to the found party. The automatic transfer means 74 identifies the speaker by the voice of the caller, recognizes the destination of the caller's voice, and searches for a pre-registered number corresponding to the destination.
If you make a call with the searched number and if there is a response to the call with the searched number, you will be notified of the name of the caller and the partner of the searched number with a registered synthetic voice, and you will be asked to act as an agent. Reference numeral 75 denotes an automatic registration means for automatically recording voice data of the caller and adding the caller as a registrant in the case of a caller whose voice is not registered in advance.

The operation of the automatic intermediary means 71 of the computer 7 configured as described above will be described with reference to the flowchart of FIG. 3, the display diagram of FIG. 4 and the data diagram of FIG.

First, an application program relating to automatic agency is executed (S1). Then, a display as shown in FIG. 4 appears on the screen of the display device of the extension terminal (multifunctional telephone) 6 of the user (sender or receiver) (S
2). The user can give instructions on the screen or the voice synthesizer 12
According to the guidance by the synthetic voice from, the instruction regarding the intermediary in normal time and in the absence is pressed by pressing a specific key of the extension terminal 6 or uttered like "transfer 102" and input from a microphone (not shown). To the computer 7 (S3). The computer 7, which has input the key information or the vocalization information as the instruction content, saves the instruction content in the external storage device 8 in a predetermined format (S).
4).

Next, it is judged whether or not the input of the transfer instruction is completed (S5), and when it is judged that the input is completed, the state of waiting for the incoming call to the extension terminal 6 of the user is set (S6). If it is determined that the processing has not ended, the processes of steps 3 and 4 are repeated. When it is determined in step 6 that there is an incoming call at the user's extension terminal 6, the process proceeds to step 7. In the state of waiting for an incoming call to the extension terminal 6 (S6), although not shown in FIG.
It is possible to inform the computer 7 that the user is absent by pressing a predetermined button.

In step 7, first, the guidance by the synthesized speech from the speech synthesizer 12 (for example, "This is ○○
The section is △△. Guidance), and wait for the caller to speak. When the caller speaks, the voice is taken into the voice recognition device 11 and a speaker identification process is performed (S8).

Next, the speaker identification processing in step 8 will be described with reference to the flowchart shown in FIG. 5 and the block diagram of the voice recognition device 11 shown in FIG. The voice signal of the caller input through the telephone line is the bandpass filter 13
After being filtered by, the signal is sampled and quantized by the A / D converter 14 (S21). The acoustic analysis unit 15 cuts out a voice section from the input voice signal (S22),
The voice signal is converted into a series of parameters (parameters such as pitch period and average amplitude) that more clearly represent the characteristics of the speaker (S23). Next, the pattern conversion unit 16 configures a pattern vector by discretizing or associating time with a reference pattern (S24). Next, the determination unit 1
In step 7, the distance between the previously registered standard pattern vector of each registrant and the pattern vector is calculated (S2
5) Identifying which speaker the speaker is registered in the system (speaker identification), or verifying whether the speaker matches the speaker with the speaker ID input separately (speaker verification) ( S2
6).

Next, returning to FIG. 3, the processing is branched based on the result of the speaker identification processing or the speaker verification processing (identification processing) (S9). In the identification processing, if the person is found among the voice registered persons, the process proceeds to step 10, and if not, the process proceeds to step 12. In step 10, it is judged whether or not there is an intermediary instruction for the person in question, and if there is an instruction, the process proceeds to step 11, and if there is no instruction, the process proceeds to step 12. In step 11,
The agency process is performed based on the agency instruction. The agency process in step 11 and the agency process by the knowledge base in step 12 (that is, the agency process according to the regulation of storage in the storage device 8) are the agency process means 72 and the automatic transfer means 7 in FIG.
The details of these processes will be described later with reference to the flowcharts of FIGS. 8 to 10. When the intermediary process is completed in step 11, the process returns to step 6 to enter the state of waiting for an incoming call again. Step 12
Then, the processing to be taken is determined according to the rules (see FIG. 7) regarding the telephone intermediary stored in the external storage device 8,
The processing described later is performed. Then, the process returns to step 6 to enter the state of waiting for an incoming call again.

The relationship between each step and each means will be described. Steps 1 to 10 are automatic intermediary means 71.
, And steps 11 and 12 show the processing operations in the intermediary processing means 72 and the automatic transfer means 73 as described above.

Next, details of the intermediary process in step 11 will be described with reference to the flowchart of FIG. As the processing contents of the telephone transfer, as shown in step 31 of FIG. 8, "the user (the receiver here) answers", "transfer to another telephone (other place)" and "general" There are three possible cases: "Announce to the caller to leave a message like an answering machine."

First, as described above, in step 31, the process is branched depending on the contents of the agency instruction. In the case of the instruction "the user answers as it is", the calling bell is sounded (S32) and the user waits for off-hook (S33). If you do not go off-hook within a certain period of time, or if the user has made absent settings, the computer 7 instructs the PBX 1 to automatically connect the line and provide guidance to the caller with a synthesized voice prepared in advance. (For example, guidance such as "I'm sorry, I can't answer right now. Please call again or leave a message after the dial tone.")
(S34). Then, after a certain period of time, a dial tone is emitted to start recording a message of the caller, and when the caller goes on-hook, the PBX 1 is made to blink the LED of a predetermined key of the extension terminal 6 of the user. To inform the user that there is a message, and the process ends (S35). If the user goes off-hook within a certain time in step 33, the user responds, and the processing in the computer 7 ends.

If the transfer instruction is "transfer to another telephone (other place)", the process proceeds to step 36. In step 36, the guidance (for example, "I am in another place.
We will forward your call. Please wait for a while "
Such as guidance), and the extension terminal 6 is temporarily
After the call is put on hold, in step 37, the telephone number (102 in FIG. 4) registered by the user is automatically called to perform the transfer process. When the transfer destination telephone answers, a message such as "I want to call Mr. △△" is played by the synthesized voice, and the user is requested to act as an agent (S38).

In the case of an intermediary instruction of "announce the message to the caller to leave a message like a general answering machine", the processing in step 34 and thereafter is performed.

The relationship between each step and each means will be described here. Steps 31 to 35 are the automatic intermediary means 7.
1 shows the processing operation in the agency processing means 72, and steps 36 to 38 show the processing operation in the automatic transfer means 73.

Next, the telephone transfer process according to the transfer rule in the case of the default (no transfer instruction) in step 12 will be described in detail with reference to the data diagram of FIG. 7 and the flowcharts of FIGS. 9 and 10. First, it is determined whether or not the sender's name, group name, or the like corresponds to the headline in the agency rule (S41). When it is determined that there is a corresponding item, the process proceeds to step 42,
If it is determined that there is no corresponding item, the process proceeds to step 49. In step 42, the instruction in the agency rule as shown in FIG. 7 is read out, and in step 43, the processing based on the instruction is performed. For example, when the caller is Mr. Mori, according to the description of FIG. 7, since the highest priority is given (because the Priority of Title is "MORI" is "A"), the ringing of the ringing call and the synthesized voice are performed. By saying, "I'm calling from Mr. Ms. Mori.
Please send a message that calls attention such as "Please take it" (S
44). Next, it is determined whether or not there is a response (S45). If it is determined that there is a response after waiting for a certain period of time, this process ends, but if it is determined that there is no response, the synthesized voice will still say, "Sorry. I'm absent now. Do you want to search?" Is asked to determine whether the caller's predetermined key press or voice response is affirmative or negative (S46). When it is determined that the response is affirmative, the process proceeds to step 49, and when it is determined that the response is negative, in step 47, a guidance is given to leave a message by synthetic voice again, and step 48.
Then, the message is recorded, and the process ends.

In step 49 and the subsequent steps, automatic call processing to a potential destination is performed. The possible turnaround destination is the telephone number of the transfer destination that is input to the computer 7 in advance by the user (the receiver here).

In step 49, a message such as "I am looking for △△, please wait for a while" is sent by the registered synthetic voice, and the caller is made to wait. Next, the list of possible turnaround destinations is read from the external storage device 8 (S50). Next, it is determined whether the list is over (S51). If it is determined that the list is not over, the telephone number is taken out from the list and the number is called (S52). Next, it is determined whether or not the other party has responded (S53). If it is determined that the response has been made, a question such as "Is there a △△ of XX section over there?" Is asked by the synthesized voice (S5).
4) Wait for a reply by a predetermined button or voice of the responder (S55). When it is determined in step 53 that no response is made, the process returns to step 51 and it is determined whether or not the list is completed.

When it is determined in step 55 that the answer is affirmative, the call is connected by the transfer process and the process is terminated (S).
56). If the answer is negative, the call is disconnected (S57) and the process returns to step 51.

When it is determined in step 51 that the list is at the end, a message such as "Sorry. △△ was not found" is sent to the caller by the registered synthetic voice (S58). A message such as "Please send after the dial tone if there is a message" is played (S59). Next, it is determined whether the caller has hung up (S60). When it is determined that the caller has hung up, the process ends at that point. If it is determined that the call has not been hung up, a message is recorded and the process ends (S
61).

Now, the relationship between each step and each means will be described. Steps 41 to 48 are automatic intermediary means 7.
1 shows the processing operation in the agency processing means 72, and steps 49 to 61 show the processing operation in the automatic transfer means 73.

Although the functions provided in this embodiment are constructed on the computer 7, they may be constructed on the computer 10, and the processing operation in this case is similar to that of the computer 7. . Further, in the present embodiment, the instruction regarding the agency is based on the application program. However, the instruction regarding the agency can be input from the computer 7 in advance and stored in the external storage device 8 independently of the application program. In this case, the display of FIG. 4 is not made.

As described above, according to this embodiment, the speaker is identified by the voice of the caller, a predetermined word group for guidance or the like is generated for the caller, and the user (receiver ) And the intermediary rules prescribed for storing in the external storage device 8 are used to automatically perform the intermediary work, so that the telephone intermediary work that is performed by a human secretary can be performed automatically. The convenience for the person can be further enhanced. In addition, if the person who is the automatic agent is not available, P to the extension telephone number registered in advance.
Since the call is automatically made from the BX1, the other party is searched by the synthesized voice, and the incoming call is transferred from the PBX1 to the found party, so that the transfer of the telephone can be automated.

(Embodiment 2) A private telephone system according to a second embodiment of the present invention and an information processing method of the private telephone system will be described below with reference to the drawings.

FIG. 11 is a flow chart for explaining the processing operation of the private telephone system according to the second embodiment of the present invention, and shows the processing operation in the automatic transmission intermediary means 74 of FIG.

In FIG. 11, first, the name of the other party spoken by the caller is taken into the voice recognition device 11 (S71),
The telephone number of the other party (callee) is searched, and the caller is identified by speaker identification (S72). Next, the PBX 1 is instructed to call the telephone number of the other party (S73), and the response of the other party is waited for (S74). If it is judged that the other party responded, the registered synthetic voice tells the name of the sender and the other party, and a message requesting an intermediary (for example, "I'm calling Mr. ○○ from Mr. ○○. Please send a message such as "Please") (S7
5). When it is determined in step 74 that the other party does not answer even after calling for a certain period of time, a message such as "No. □□ is not answered. Do you want to redial?" Is sent to the caller (S76). , The word of the user's response as to whether or not to redial is recognized (S77). If it is recognized that the redial is made after a certain period of time, the redial is performed (S78), and if it is recognized that the redial is not performed, the process ends.

As described above, according to this embodiment, the speaker is identified by the voice of the caller, the callee's utterance destination is recognized as a word, and the number registered in advance corresponding to the callee is searched. , If you make a call with the searched number and if there is a response to the call with the searched number, the name of the caller and the partner of the searched number will be announced by the registered synthetic voice and you will be asked to act as an agent. In addition, it is possible to automatically perform a telephone intermediary work, which is performed by a human secretary, and further improve convenience for the user.

(Third Embodiment) A private telephone system according to a third embodiment of the present invention and an information processing method of the private telephone system will be described below with reference to the drawings.

FIG. 12 is a flow chart for explaining the processing operation of the private telephone system according to the third embodiment of the present invention, and shows the processing operation in the automatic registration means 75 of FIG.

In FIG. 12, a guidance (for example, "This is XX section. Please give me your name") from the voice synthesizing device 12 is played and the caller's voice is awaited (S81). When the caller speaks, the voice is taken into the voice recognition device 11 and a speaker identification process is performed (S82). The process branches based on the result of the identification process (S83). If it is determined in the identification process that there is a corresponding person among the registrants, the processing from step 10 onward in FIG. 3, that is, the processing in the agency processing means 72 is performed (S8).
4). When it is determined in the identification process that there is no corresponding person among the registrants, the voice data uttered by the caller is stored in the storage device 8, and the feature amount of the voice necessary for the speaker identification process is extracted from the voice data. , Storing with name (S8
6). After that, the processing from step 12 onward in FIG. 3, that is, the processing in the agency processing means 72 is performed (S8).
7).

As described above, according to the present embodiment, in the case of a caller whose voice is not registered in advance, the utterance data of the caller is automatically recorded, the caller is added as a registrant, and thereafter, the caller acts as an agent. Since the processing is performed, it is possible to automatically perform the telephone intermediary work similar to that performed by a human secretary even for unregistered callers, and further improve convenience for users. be able to.

[0045]

As described above, the present invention is connected to an automatic private branch exchange, to the automatic private branch exchange through a dedicated interface, to a computer having a voice input / output interface, and to the computer and the voice input / output interface. A voice recognition device that identifies a speaker by the voice of the user, a voice synthesis device that is connected to a computer via a voice input / output interface, and generates a predetermined word group for guidance or the like to the user, A storage device having a database accommodating a group of rules concerning telephone intermediation, an automatic intermediary means for automatically performing an intermediary for an incoming call according to an instruction given by a user or an intermediary intermediary rule for storing the storage device, and an automatic intermediary If the other party is absent, use the automatic private branch exchange to the extension telephone number registered in advance. By making a call dynamically, searching for the other party by the synthesized voice, and by providing an automatic transfer means for transferring the incoming call from the automatic private branch exchange to the found party, the speaker can be identified by the voice of the caller in advance. It is possible to generate a set of words for guidance such as guidance for the sender, and to automatically perform the transfer according to the instructions given by the receiver and the specified transfer rules, so that a phone call made by a human secretary. It is possible to realize an on-site telephone system that can automatically perform an intermediary work and further improve convenience for users. In addition, when the other party who is the automatic agent is absent, the PBX1 automatically calls the extension telephone number registered in advance, the other party is searched by the synthesized voice, and the incoming call is transferred from the PBX1 to the found party. As a result, it is possible to realize a local telephone system that enables automation of telephone intermediary even during transfer.

Further, the automatic private branch exchange, the computer, the voice recognition device, the voice synthesizer, the storage device, speaker identification by the voice of the caller, and the caller's destination Recognize words, search for a pre-registered number corresponding to the destination, and call with the searched number,
When there is a response to the calling of the searched number, the call is sent by providing an automatic calling agent that tells the name of the sender and the partner of the searched number by the registered synthetic voice and requests an agent. The speaker's voice is used to identify the speaker, the caller's utterance destination is recognized as a word, the pre-registered number corresponding to the callee is searched, and the call is made using the searched number. When there is a response, it is possible to request the agent by telling the name of the sender and the partner of the searched number with the registered synthetic voice.
It is possible to realize a private telephone system that can automatically perform a telephone intermediary work that is performed by a human secretary and can further improve convenience for users.

Further, the automatic private branch exchange, the computer, the voice recognition device, the voice synthesizing device, the storage device, and the caller whose voice is not registered in advance automatically call the caller's A caller whose voice is not registered in advance by providing automatic registration means for recording vocalization data and adding the caller as a registrant, and an intermediary processing means for performing an intermediary process based on the registration data of the automatic registration means. In this case, the voice data of the caller can be automatically recorded, the caller can be added as a registrant, and the intermediary process can be performed after that. It is possible to realize an on-premise telephone system that can automatically perform such a telephone intermediary work and further improve convenience for users.

Furthermore, an automatic intermediary process for automatically performing an intermediary for an incoming call according to an instruction given by a user or an intermediary intermediary rule prescribed in the storage of the storage device, and if an automatic intermediary partner is absent, it is registered in advance. The automatic private branch exchange automatically calls the specified extension telephone number, searches for the other party by the synthesized voice, and has an automatic transfer step of transferring the incoming call from the automatic private branch exchange to the found party. The speaker is identified by voice, a predetermined word group for guidance is generated for the caller,
Since the agency can automatically perform the agency according to the instructions given by the recipient or the prescribed agency rule, it is possible to automatically perform the telephone agency work as a human secretary does, and to the user. It is possible to realize an information processing method of a private telephone system that can further improve convenience. In addition, when the other party who is the automatic agent is absent, the PBX1 automatically calls the extension telephone number registered in advance, the other party is searched by the synthesized voice, and the incoming call is transferred from the PBX1 to the found party. As a result, it is possible to realize an information processing method of a local telephone system which enables automation of telephone intermediation even in transfer.

Furthermore, the caller's voice is used to identify the speaker, the caller's uttered destination is word-recognized, a pre-registered number corresponding to the callee is searched, and the searched number is called to make a search. If there is a response to the calling of the number, the voice of the caller will be sent by having an automatic call transfer process that tells the name of the caller and the partner of the searched number using the registered voice The speaker is used to identify the caller, the caller's utterance destination is recognized as a word, the pre-registered number corresponding to the callee is searched, and the call is made using the searched number. If there is, it is possible to request the agent by telling the name of the sender and the other party of the searched number by the registered synthetic voice, so that the telephone agent work like a human secretary automatically performs. Can Also with respect to the user can realize an information processing method private telephone system which can improve convenience more.

Furthermore, in the case of a caller whose voice is not registered in advance, the caller's utterance data is automatically recorded and an automatic registration step of adding the caller as a registrant, and an intermediary process based on the registration data. By having an intermediary processing step to be performed, in the case of a caller whose voice is not registered in advance, the vocalization data of the caller is automatically recorded, the caller is added as a registrant, and then the intermediary process is performed. Because it can be done, it is possible to automatically perform telephone transfer work similar to that performed by a human secretary even for unregistered callers, and to further enhance convenience for users. The information processing method of the telephone system can be realized.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a private telephone system according to an embodiment of the present invention.

FIG. 2 is a functional block diagram showing a function realizing means of a computer that constitutes a private telephone system according to an embodiment of the present invention.

FIG. 3 is a flowchart for explaining the operation of the automatic intermediary means of FIG.

FIG. 4 is a display diagram showing a screen of a display device of the multi-function telephone of FIG.

5 is a flowchart for explaining the operation of the acoustic recognition device in FIG.

6 is a block diagram showing the speech recognition apparatus of FIG.

FIG. 7 is a data diagram showing data used when there is no transfer instruction.

FIG. 8 is a flowchart for explaining the operation of the agency processing means and the automatic transfer means of FIG.

FIG. 9 is a flowchart for explaining the operation of the agency processing means and the automatic transfer means of FIG.

FIG. 10 is a flowchart for explaining the operation of the agency processing means and the automatic transfer means of FIG.

FIG. 11 is a flowchart for explaining the operation of the automatic transmission intermediary means of FIG.

12 is a flowchart for explaining the operation of the automatic registration means of FIG.

FIG. 13 is a configuration diagram showing a conventional private telephone system.

[Explanation of symbols]

 1 PBX 2 Station line 3 Public telephone network 4 Extension 5 Extension phone 6 Multi-function telephone 7, 10 Computer 8 External storage device 9 Transmission path 11 Speech recognition device 12 Speech synthesizer 13 Band filter 14 A / D converter 15 Acoustic analysis unit 16 pattern conversion unit 17 determination unit

Claims (6)

[Claims] 1. An automatic private branch exchange, a computer that is connected to the automatic private branch exchange through a dedicated interface, and has a voice input / output interface, and that is connected to the computer through the voice input / output interface, depending on the voice of the user. A voice recognition device for identifying a speaker, a voice synthesizer connected to the computer through the voice input / output interface, for generating a predetermined word group for guidance or the like to the user, and a telephone intermediary. A storage device having a database accommodating a group of rules relating to the incoming call, an automatic intermediary means for automatically performing an intermediary for an incoming call according to an instruction given by a user or an intermediary intermediary rule for storing the storage device, and an automatic intermediary partner. If you are absent, call the registered internal telephone number to the automatic private branch exchange. Et automatically calling, synthesized speech by looking for the other party, private telephone system characterized by comprising an automatic transfer means for transferring an incoming call from the automatic private branch exchange to the other party the found. 2. An automatic private branch exchange, a computer that is connected to the automatic private branch exchange through a dedicated interface, and has a voice input / output interface, and that is connected to the computer through the voice input / output interface, and according to the voice of the user. A voice recognition device for identifying a speaker, a voice synthesizer connected to the computer through the voice input / output interface, for generating a predetermined word group for guidance or the like to the user, and a telephone intermediary. Storage device having a database accommodating a group of rules regarding the caller, speaker identification is performed by the voice of the caller, word recognition is performed on the callee's utterance destination, and a preregistered number corresponding to the callee is searched. , If you make a call with the searched number and if there is a response to the call with the searched number, Private telephone system characterized by comprising an automatic calling intermediary means to request the agency telling the name of the person the caller and the retrieved number by voice. 3. An automatic private branch exchange, a computer that is connected to the automatic private branch exchange through a dedicated interface, and has a voice input / output interface, and is connected to the computer through the voice input / output interface, and according to the voice of the user. A voice recognition device for identifying a speaker, a voice synthesizer connected to the computer through the voice input / output interface, for generating a predetermined word group for guidance or the like to the user, and a telephone intermediary. Storage device having a database accommodating a group of rules relating to the above, and automatic registration means for automatically recording the utterance data of the caller in the case of a caller whose voice is not registered in advance and adding the caller as a registrant. And an intermediary processing means for performing intermediary processing based on the registration data of the automatic registration means. Private telephone system according to claim. 4. An automatic intermediary process for automatically performing an intermediary for an incoming call according to an instruction given by a user or an intermediary intermediary rule prescribed in the storage of the storage device, and if an automatic intermediary partner is absent, pre-register. An automatic transfer step of automatically calling the specified extension telephone number from the automatic private branch exchange, searching for the other party by synthetic voice, and transferring an incoming call from the automatic private branch exchange to the found other party. A characteristic information processing method for a local telephone system. 5. The speaker is identified by the voice of the caller, the callee's utterance destination is word-recognized, a preregistered number corresponding to the callee is searched, and the call is made with the searched number. When there is a response to the transmission of the searched number, it has an automatic transmission mediation step of telling the name of the caller and the partner of the searched number by a registered synthetic voice and requesting the agency. Information processing method for a local telephone system. 6. In the case of a caller whose voice is not registered in advance, an automatic registration step of automatically recording the caller's utterance data and adding the caller as a registrant, and an intermediary based on the registration data. An information processing method for a private telephone system, comprising an intermediary processing step for performing processing.