JPH0287757A - Telephone automatic reply recognition device - Google Patents

Abstract

PURPOSE:To attain write to a storage device via a telephone line without direct operation of the device and to facilitate the maintenance by providing a signal line changeover means as the maintenance means for a voice data, writing a signal of a telephone line therein and leading it to a voice analysis section required at the maintenance. CONSTITUTION:The major point of providing a telephone manual reply recognition device 3 resides in the provision of a signal switch section 37 switching a flow of a voice signal from a line control section 31 to a voice recognition section 34 or a voice analysis section 32 between the line control section 31 and the voice recognition section 34. Thus, the signal line is at first connected to the recognition section 34 and when an instruction comes from the control section 35, the signal is switched automatically to the voice analysis section 32. At the maintenance of the data through the constitution above, in the presence of a voice input signal for the maintenance from the telephone line 5, the switching section 37 is switched and the voice data is inputted to a magnetic storage device 36 via the analysis section 32 to apply revision registration based thereupon and when the job is finished, the state is restored to the initial state.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to an automatic telephone response recognition device that modifies the voice data maintenance method.

[Prior Art] FIG. 2 is a block diagram showing an example of the configuration of a conventional automatic telephone response recognition device. The user operates telephone 1 to
For example, a series of predetermined code numbers is divided into words and uttered. The uttered voice signal is input to the automatic telephone response recognition device 3 through the exchange 2. At this time, the line control unit 3 becomes the input/output stage of the automatic telephone response recognition device 3.
1 connects the telephone line 5 to the automatic telephone response recognition device 3 according to instructions from the control section 35, and electrically connects the voice response section 33 and the voice recognition section 34. Voice recognition section 34
recognizes the code number uttered by the user word by word, and notifies the host 4 of the result through the control unit 35. The voice data to be provided to the user is registered in the magnetic storage device 36, and as shown in FIG. 3, the voice data is assigned a voice number and converted into a digital signal.

Here, when performing voice synthesis according to an instruction from the upper host 4, the control unit 35 retrieves the voice data corresponding to the voice number sent from the upper host 4 from the magnetic storage device 36, and transfers it to the voice response unit. Send to 33. Voice response section 3
3 converts the sent voice data from digital to analog and passes it through the line control unit 31, the exchange 2, and the telephone 1.
Sends sound□voice to the user. Reference numeral 37 denotes an operation section of the device 3, which includes, for example, a display section 371 that displays the operating status of the device, a power switch 372, a recovery switch 373, and a switch 37 for selecting and starting maintenance functions, as shown in FIG.
4 are provided.

In addition, in order to register or change audio data, use the operation section 37 to put the device 3 into a maintenance state, and then use the operation section 37.
This is done by inputting a voice number from the microphone 38 and then inputting voice from the microphone 38. The input audio is
The signal is sent to the voice analysis section 32, where it is analog/digital converted and sent to the control section 35. The control unit 35 assigns a voice number to the received voice data and stores it in the magnetic storage device 36.
Change the registered data with the voice number registered or input.

Furthermore, confirmation of the registered voice data is performed by inputting the voice number of the voice to be confirmed from the operation unit 37, and as in the previous voice synthesis, the control unit 35 inputs the voice data from the magnetic memory bag 'ff 36. The data is collected, sent to the voice response section 33, and confirmed by the voice emitted from the speaker 39.

Note that at this time, the voice response section 33 is connected only to the speaker 39, and the connection with the line control section 31 is cut off.

[Problems to be Solved by the Invention] However, in order to perform maintenance work such as registering and changing the magnetic storage device containing audio data with the device having the above configuration, it is necessary to perform maintenance work while actually operating the device. Therefore, in order to perform maintenance on a plurality of devices, it is necessary to secure a plurality of maintenance personnel or to have one maintenance personnel visit each device and perform maintenance work.

For this reason, as shown in Figure 5, the head office, which centrally manages information, has multiple branch offices that are equipped with automatic telephone response recognition devices and that actually provide services to users. →If the system has a system configuration where there are many users receiving Novice service, for example, if the system provides services such as fresh food prices and stock information that change every day, it may be necessary to maintain multiple devices at the same time. However, the apparatus having the configuration shown in FIG. 2 described above has problems such as high labor costs and the inability to provide prompt service.

The purpose of this invention is, as mentioned above, to maintain the audio data of multiple devices, it is necessary to secure multiple maintenance personnel. In the case of
An object of the present invention is to provide an automatic telephone response recognition device with excellent maintainability.

[Means for Solving the Problems] The automatic telephone answering recognition device of the present invention guides a code signal from a telephone line to a signal recognition unit during speech synthesis, reads out voice data based on the recognition result from a storage device, and performs a telephone call. In an automatic telephone answering recognition device that synthesizes and outputs the voice data to the telephone line and writes the voice data to the memory device via the voice analysis section during data maintenance of the storage device, the flow of the input signal from the telephone line is analyzed by the signal recognition section. and control means for switching the signal path switching means in response to a signal from the telephone line during data maintenance and writing voice data from the telephone line to the storage device via the voice analysis section. It is something.

[Function] During data maintenance, when a maintenance worker inputs a maintenance signal such as voice from the telephone line into the automatic telephone answering recognition device, the signal path switching means is switched and the maintenance function is activated, and the signal path that was previously guided to the signal recognition unit is switched. The input signal from the telephone line is guided to the voice analysis section.

Here, when the maintenance worker inputs voice data for maintenance into the automatic telephone response recognition device through the telephone line, the voice data is directly written into the storage device via the voice analysis section.

Therefore, audio data can be written for maintenance while the user is present, without directly operating the device.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the automatic telephone response recognition device of the present invention. In the figure, the same parts as those in the conventional example shown in FIG. 2 are denoted by the same reference numerals, and the explanation thereof will be omitted.

The main difference from FIG. 2 is that the flow of the audio signal from the line control unit 31 is controlled between the line control unit 31 and the voice recognition unit 34.
The point is that a signal road exit J material section 37 that can be switched to the voice recognition section 34 or the voice analysis section 32 is provided. Initially, the signal path is connected to the voice recognition section 34, but can be freely switched to the voice analysis section 32 by command from the control section 35.

Further, during data maintenance, the control means switches the signal path switching section 37 according to the voice input signal from the telephone line 5 and writes the voice data from the telephone line 5 to the magnetic storage device 36 via the voice analysis section 32. Figure 6, which will be described later.
~ It can be configured by a microcomputer programmed as shown in FIG. 9, and specifically realized by the control section 35.

Next, a method of operating the automatic telephone response recognition device according to the present invention will be explained. FIG. 6 is a flowchart of initial control in this invention. When the user calls the device 3 using the telephone 1 and the line connection by the line control section 31 is completed, the device 3 sends initial guidance requesting service content to the user (S6L).

Since the signal path switching unit 37 is initially connected to the voice recognition unit 34, the answer from the user is input to the voice recognition unit 34, and the result is notified to the control unit 35 (S62).

The control unit 35 determines whether or not this notification is a request for maintenance work on the audio data (863), and if it is not a request for maintenance work, starts normal service (S64).
). Here, if the request is for maintenance work, next, in order to determine whether the user is a maintenance worker, a guidance requesting a password is sent to the user name (S65).
. Similarly to the above, it is important to recognize responses from users56.
6) The control unit 35 determines whether the result matches a predetermined number 567), and if the password from the user is correct, the voice data maintenance work is started (S68). . Note that if the password is incorrect, normal service is started (864).

FIG. 7 is an initial control flow of voice data maintenance work. When maintenance work is started, guidance for confirming the work content is first sent from the voice response unit 33 to the maintenance personnel using the telephone 1 in accordance with an instruction from the control unit 35 (S71). The answer from the maintenance worker is recognized by the voice recognition section 34, and the result is sent to the control section 35.
(S72). The control unit 35 determines from this notification whether the work content is maintenance work completion (373), audio data registration/change (S74), or audio data confirmation (S75), and then shifts control to the corresponding work.

The control flow in FIG. 8 shows control when the work content is registration/change of audio data. First, guidance is sent to maintenance personnel requesting the voice number of the voice data to be registered/changed (S81).

The answer from the maintenance worker is recognized by the voice recognition section 34, and is notified to and stored in the control section 35 (S82). Change here/
It is determined whether the registration work has been completed (883), and if this notification matches the predetermined work completion number, the work is ended (S84). Next, after sending out guidance prompting voice input (S85), the control unit 35
sends a command to the signal path switching unit 37 to switch the signal path that has been connected to the voice recognition unit 34 to the voice analysis unit 32 (386). Signal path switching unit 3 that received the command
7 switches the signal path to the voice analysis section 32, and thereafter, the signal from the line control section 31 is sent to the voice analysis section 32.

The voice analysis unit 32 has a function of monitoring silent time, and
The input signals are analog/digital converted and stored in memory until a silent time of 0 seconds is detected, and after the silent time is detected, the controller 35 is notified as audio data (S87.588). Upon receiving the notification, the control unit 35 changes and registers the voice data in the magnetic storage device 36 corresponding to the previous voice number (889), and reconnects the signal path switching unit 37 to the voice recognition unit 34. After (S90)
, return to the initial state of voice data registration/change.

Similarly, the audio data confirmation work is performed according to the control flow shown in FIG. A voice number input request guidance is sent out (S91), and after the voice number from the maintenance worker is recognized (S92), it is determined whether the confirmation work has been completed (893). If not, the control unit 35 retrieves audio data from the magnetic storage device 36 and sends this audio data to the audio response unit 33. Here, the control section 35 sends the voice data sent to the voice response section 33 not to the speaker 39 but to the line control section 31 (894), and delivers the voice to the maintenance personnel via the telephone 1.

As described above, according to this embodiment, the magnetic storage device 3
6 can be performed using the telephone line 5 without actually operating the operation unit 37 of the device 3, so the system is configured to provide services such as prices and stock information of produced foods to users. Magnetic storage device W36 of a plurality of devices 3
Even when performing maintenance, only one maintenance person is required because the work can be done from one location, and the maintenance work can be carried out economically and quickly.

Although the above embodiment describes only an automatic telephone response recognition device capable of voice recognition, the present invention can also be applied to a device having a multi-frequency signal wave or Oribe (PB data) instead of the voice recognition section 34. It is possible to apply.

[Effects of the Invention] According to the present invention, the signal path switching means is provided so that the telephone line signal can be written and guided to the voice analysis section necessary for maintenance, so that the telephone line can be switched without directly operating the device. Since data can be written to the storage device via the , maintenance work becomes extremely easy.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the automatic telephone answering recognition device of the present invention, FIG. 2 is a block diagram showing an example of a conventional automatic telephone answering recognition device, and FIG. 3 is a block diagram showing an example of the conventional automatic telephone answering recognition device. A configuration diagram of voice data in a magnetic storage device, FIG. 4 is a configuration diagram of the operating section of an automatic telephone response recognition device, FIG. 5 is a configuration diagram of an example of a European system of an automatic telephone response recognition device, and FIG. 6 is a diagram of the system according to the present invention. Example initial control flowchart q-t-1 FIG. 7 is an initial control flowchart of voice data maintenance work according to an example of the present invention, FIG. 8 is a control flowchart of voice data registration/change work according to an example of the present invention, and FIG. It is a control flowchart of the audio data confirmation work of the example of this invention. 3...Telephone automatic answer recognition device, 5...Telephone line, 3
2... Voice analysis unit, 34... Voice recognition unit as a signal recognition unit, 35... Control unit, 36... Magnetic storage device, 37... Signal path switching unit Patent applicant Oki Electric Industry Co., Ltd. Figure 4 Figure 5 Initial control of this embodiment 7a = ft- Figure 6 Control of registration/change work chart Figure 8 Initial control of maintenance work chart Figure 7 Control of confirmation work 7 tochir ) Figure 9

Claims (1)

[Claims] During voice synthesis, a code signal is guided from a telephone line to a signal recognition section, and voice data based on the recognition result is read from a storage device and synthesized and output to the telephone line, and when data maintenance of the storage device is performed. , an automatic telephone response recognition device that writes voice data to a storage device via a voice analysis section, comprising a signal path switching means for switching the flow of input signals from the telephone line from the signal recognition section to the voice analysis section; An automatic telephone response recognition device comprising: control means for switching the signal path switching means in response to a signal from the line and writing voice data from the telephone line to the storage device via a voice analysis section.