JPH04100470A - Telephone exchange system - Google Patents

Abstract

PURPOSE:To automate telephone call answering and message recording when a called user of a telephone set is absent by providing a speech path exchanging section, audio responding section, voice recognizing section, voice accumulating section, and controlling section. CONSTITUTION:When a incoming call automatically arrives from a telephone line through a telephone line connecting terminal 11, a speech path exchanging section l outputs the telephone line number of the incoming call to a controlling section 5 as an incoming signal and sets buses between the telephone line of the incoming call and an audio responding section 2, voice recognizing section 3, and voice accumulating section 4. The controlling section 5 outputs an audio message to be used for incoming calls in the audio responding section 2 to the telephone line of the incoming call in response to the incoming signal from the section l. When the section does not output an off-hook signal within prescribed time, the section 5 outputs an absence message. When the handset of the telephone set connected to a telephone set connecting terminal 12 is lifted up, the section 5 checks whether or not any message exists to the user of the telephone set and, when a message exists, outputs reproduction controlling data to the voice accumulating section 4. Then voices (message) read out from the section 4 are conveyed to the user of the telephone set when the user lifts up the handset.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a telephone exchange system equipped with a voice recognition function, a voice response function, and a voice storage function.

[Conventional technology]

In traditional telephone switching systems, a call from a telephone line is answered by an operator, the caller confirms the name of the recipient, and then the operator confirms the presence of the recipient;
When he was present, he operated the incoming call connection to the recipient. In addition, when the recipient is absent, the sender is notified that the recipient is absent, and if there is a message from the sender, the message is recorded.
A memo containing a message was delivered to the recipient's desk.

[Problem to be solved by the invention]

The conventional telephone switching system described above requires the intervention of an operator, which increases operator labor costs.Also, because the number of operators is limited due to this cost constraint, as the number of lines increases, callers receive more and more calls. This not only increases the amount of time it takes for the operator to connect the telephone, but also has the drawback that incoming calls cannot be accepted when the operator is absent, such as at night.

[Means to solve the problem]

The telephone switching system of the present invention includes a line switching section that performs exchange connections between N telephone line connection terminals and M telephone connection terminals based on input connection control data; a voice response unit that synthesizes voice from data and outputs it to the line switching unit; a voice recognition unit that inputs the voice output from the line switching unit, performs voice recognition, and outputs recognized data; and the line switching unit a voice storage section that inputs an analog voice signal from a computer and stores a converted digital voice signal, and regenerates the accumulated digital voice signal into an analog voice signal and outputs it to the line switching section; The voice response unit includes a control unit that performs control including output control of the response data and output control of the connection control data to the line switching unit, and the voice response unit performs recording using an ADPCM encoding method.・Performs reproduction-type speech synthesis, and the speech recognition unit analyzes the speech of an unspecified number of speakers in advance and extracts the characteristics of the majority of speakers to create a standard pattern and the speech signal input during recognition. and the input cover pattern from which the features are extracted are recognized by pattern matching, and the audio storage section performs recording/playback type storage/playback using an ADPCM encoding method.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention.

This telephone switching system includes a voice response section 2, a voice recognition section 3, a voice storage section 4, a line switching section], and a control section 5.

The voice response unit 2 synthesizes voice from the response data output by the control unit 5 and outputs the synthesized voice to the line switching unit 1. The voice response unit 2 performs recording and playback using a recording/playback type using, for example, the ADPCM encoding method. That is, the speech signal in units of words or phrases is digitized in advance using the ADPCM encoding method and stored in the memory (not shown) of the speech response section 2, and the data is read out based on instructions for response data from the control section 5. and reproduce it into an audio signal.

The speech recognition unit 3 performs speech recognition of unspecified speakers.

The voice signal output from the line switching unit 1 is recognized and recognition data is output to the control unit 5. The recognition method of the speech recognition unit 3, for example, analyzes the voices of an unspecified number of speakers in advance, extracts the characteristics of those speakers, creates a standard pattern, and then analyzes the input speech signal at the time of recognition. This is a recognition method that performs pattern matching with the input cover pattern whose features are extracted.

The audio storage section 4 digitizes the audio signal output from the line switching section 1 according to the recording control data from the control section 5, and stores it in a memory area designated by the recording control data. It also reproduces the digitized audio signal stored in the memory area specified by the reproduction control data from the control section 5 and outputs it to the line switching section 1. The audio storage unit 4 is, for example, an ADPCM
Recording and playback using a recording/playback method using an encoding method 6
The line switching unit 1 accommodates telephone line connection terminals 11 with N lines and telephone connection terminals 12 with M lines, and connects the telephone line connection terminals 11 and the telephone connection terminals 12 according to connection control data from the control unit 5. connect or disconnect the bus.

Next, the operation of this embodiment will be explained.

When an incoming call is automatically received from a telephone line via the telephone line connection terminal 11, the line switching section 1 outputs the incoming telephone line number as an incoming call signal to the control section 5, and the incoming telephone line and voice response section 2
．． Speech recognition unit 3. A path is established between the audio storage units 4.Then, the control unit 5 causes the telephone to output a calling signal via the telephone connection terminal 12 in a generally known manner. When the telephone goes off-hook, the line switching section 1 outputs this telephone line number to the control section 5 as an off-hook signal.

The control unit 5 uses the incoming call signal from the line switching unit 1 to send a voice message (for example, “This is
is. Please enter the recipient's name. Outputs response data for outputting PIP j) to the incoming telephone line. When the voice response section 2 completes outputting the voice message, the control section 5 starts time monitoring, and the voice recognition section 3 inputs recognition data to the control section 5. When the control unit 5 receives this recognition data,
The call data is output to the line switching section 1 as the line number of the telephone connection terminal 12. If the voice recognition unit 3 does not output recognition data within a predetermined time, the control unit 5 outputs a telephone line disconnection signal to the line switching unit 1.

After the control unit 5 outputs the call data to the line switching unit 1,
Time monitoring is started, and when an off-hook signal is received from the line switching section 1 within a predetermined time, connection data between the incoming telephone line and the telephone line to be connected is sent to the line switching section 1 to set up a path. If the line switching section 1 does not output an off-hook signal within a predetermined period of time, the control section 5 sends a message such as "Mr. Voice response unit 2 outputs the response data for outputting ``Please speak after ``PiJ''.
Output to. When the voice response section 2 completes the output of the recipient absent message, the control section 5 outputs recording control data to the voice storage section 4. When a predetermined time elapses or when the caller puts the telephone on-hook, the control section 5 outputs a line disconnection signal to the line exchange section 1 and also outputs an instruction to stop voice storage to the voice storage section 4.

When the telephone connected to the telephone connection terminal 12 goes off-hook, the line switching section 1 sends the telephone line number to the off-hook.
This is output as a hook signal to the control unit 5, and the control unit 5 checks whether there is a message for the off-hook telephone user.

Here, if there is a message, the control unit 5 outputs playback control data to the audio storage unit 4, and the audio read from the audio storage unit 4 (
message) is transmitted to the off-hook telephone user. When all the messages have been reproduced or there is no message, the control section 5 checks the unused telephone line number of the telephone line connection terminal 11 and connects the open path between this telephone line and the telephone line.

When the control section 5 receives an on-hook signal for one or both of the telephone line or the telephone line from the line switching section 1, it sends a line disconnection signal to the line switching section to open the path between the incoming telephone line and the telephone line.

〔Effect of the invention〕

As explained above, the present invention has a circuit switching section, a voice response section, a voice recognition section, a voice storage section, and a control section, thereby automating all telephone transfers and recording of messages when the recipient is absent. This not only eliminates the need for an operator and saves costs, but also allows for mechanization to allow reception even at night.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention. ]...Line switching section, 11...Telephone line connection terminal, 1
2...Telephone connection terminal, 2...Voice response section, 3...
Voice recognition unit, 4... Voice storage unit, 5... Control unit.

Claims (1)

[Scope of Claims] 1. A line switching unit that performs exchange connections between the accommodated N telephone line connection terminals and M telephone connection terminals based on input connection control data, and input response data. a voice response unit that synthesizes voice from and outputs the synthesized voice to the line switching unit;
a voice recognition section that inputs the voice output from the line switching section, performs voice recognition, and outputs recognition data; and a voice recognition section that inputs the analog voice signal from the line switching section and stores the converted digital voice signal. an audio storage unit that reproduces the stored digital audio signal into an analog audio signal and outputs it to the line switching unit; output control of the response data to the audio response unit; and output of the connection control data to the line switching unit; 1. A telephone exchange system comprising: a control unit that performs control including control. 2. The voice response unit performs recording/playback type voice synthesis using the ADPCM encoding method, and the voice recognition unit analyzes the voices of an unspecified number of speakers in advance and extracts the characteristics of the majority of speakers. The standard pattern created by the ADP and the input pattern obtained by analyzing the audio signal input at the time of recognition and extracting its features are recognized by pattern matching, and the audio storage unit
2. The telephone exchange system according to claim 1, wherein the telephone exchange system performs recording/playback type storage/playback using a CM encoding method.