JPS58120358A - Voice response system - Google Patents

Abstract

PURPOSE:To attain required information easily and surely, by providing a function that a recipient can instruct jump and retreat in the unit of sentences or sentence group to a voice response device. CONSTITUTION:When the operator depresses a button A while a voice response device VRES outputs sentences j, p, a signal of frequency A is given to the VRES from a 4-frequency generator FGE. When a 4-frequency sound detection section DET of the VRES detects the signal of the frequency A, a control section CONT instructs the output stop of the sentences j, p and the start of output of the sentences j, p+1 to a voice synthesis section VOC. The VOC stops the output of the sentences j, p and outputs voice from the head of the sentences j, p+1. Further, when the operator depresses a button B while the VRES outputs the sentences j, p, the output of the sentences j, p is similarly stopped and the output is obtained from the head of the sentences j, p-1. Similarly, when the operator depresses a button C or D, the sentences j+1, p or j-1, p are outputted from the head.

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Technical field of the invention The present invention relates to a modification of a voice response system that responds by voice by continuously outputting sentences from a voice synthesis unit in response to a call from a telephone. There is ¥Poso.

(21) Prior art and its problems Conventionally, in voice response systems using telephone lines, the voice response device unilaterally outputs nine predetermined sentences regardless of the listener's will. When trying to get the information you want, you have to ask for information you don't need, or you have to ask for information you need, which is an inconvenience.

(3) Purpose of the Invention The present invention solves the above-mentioned inconvenience by providing a device that allows the receiver to instruct the voice response device to jump or retreat in units of sentences or groups of sentences. According to the present invention, the present invention has a telephone and a voice response device, and in response to a call from the telephone, the voice within the voice response device is activated. In a voice response system that responds to a speech machine with sentences t prepared in advance in a synthesis section, the sentences in the speech synthesis section are divided into blocks of N number of sentences, and a sentence group number and a sentence are assigned to each sentence. Add a number, 111
In response to a text skipping command or a text backward command from the handset, the control unit adds 1 to the text group number or text number and reads and outputs the text group or text numbered 9. This is accomplished by a voice response system featuring:

(5) Embodiment of the Invention FIG. 1 is a diagram showing an embodiment of the present invention, and FIG. 8 is a diagram showing the storage and output state of sentences in the speech synthesis section.

The device in the figure is a telephone, and the PGM is a 41i111 wave sound generator.

WIT is line network e VR18a voice response device, 101
7 is a network control unit, HYB is a high pre-do, DIT is a four-frequency sound detection section, VOO is a voice synthesis section, and 0OIIT is a control section.

System - When a person dials the number of the voice response device VR1B on the telephone @TIL, the network control unit in the voice response device VRE8 is called via the line network WIT. Call network control uni, )M
The OU sends a response signal back to the network MIilT.

This allows the phone to be placed horizontally and the voice response device TRI! 8 is connected. [Km1 control uni.

) NOU sends a start signal to the control unit 0ON'r @
The control unit 0ONT controls the f-voice synthesis unit WOO to sequentially output the registered sentences starting from the sentence 41-1, as shown in FIG. 2, using the activation signal ONT.

Explaining FIG. 2, closely related sentences are divided into blocks as a sentence group, and each sentence is given a sentence group number and a serial number within the sentence group.

In addition, in sentence #j, the sentence group number “j” and the serial number “1” within the group
Normally, sentences are output from sentence group 1 through hybrid HYBt- (Makoto) On the other hand, while the voice response devices VRE and B are outputting sentences j and p, for example, when a person Press button A to activate the 4-frequency sound generator FGIII.
A frequency signal from the telephone device is transmitted through the telephone line WIT to the voice response device VRIS. Voice response device V
In RmB, when the four-frequency sound detection 9DIT detects a signal of frequency M, the control unit 0ONT instructs the voice synthesis @voo to stop outputting the sentence 3e p and start outputting the sentence j * (p + 1). Speech synthesis unit VOO is text je p
Stop outputting and output audio from the beginning of sentence j, (p+1).

In this way, skipping by one sentence is achieved.

Meanwhile, the voice response device VR1B is outputting the text jt pt ratio,
When a person presses the button Bt-, the same goes for sentence 3. The output of p is stopped and one sentence je (p-1) is output from the beginning.

In this way, sentence-by-sentence regression is achieved.

Voice response device TRImB is outputting sentences j and p.

When a person presses button 0, the sentence: l(j+1), l is produced in the same way.

It is output from the destination. In this way, skipping in units of sentence groups can be realized.

If a person presses the button while the voice response device vR]l18 is outputting sentence 3e pt-, the beginning of sentence #-11,
That is, the sentence (j-1) is output from the beginning of 1. In this way, backtracking can be achieved in units of sentence groups. 5! I will clarify.

While the voice synthesis unit VOO is outputting, it notifies the control @0ONT that the voice is being output. Then, the detection unit DJi
If the signal for jumping or retreating is input manually from iτ, the control unit 0
The ONT instructs to stop the output of the audio (sentence) currently being output using the output stop signal, and sends the two-dimensional address (stored in the address storage area 3 ) for each element, + in the case of a jump
1. In the case of backwards, the operation t'1it- is performed to create a two-dimensional address of the next sentence to be output, and a new audio output instruction is given to the synthesis section.

(6) Effects of the invention As mentioned above, by providing the machine 1f@ that can give instructions for jumping over or retreating in units of sentences or groups of sentences,
You can abandon an unnecessary sentence and request the next one, or resend a sentence you missed! 1kl! ! Since you can ask for the information you need, you will be able to get the information you need easily and easily.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 8 is a diagram showing the progress state of sentences in the speech synthesis section, and FIG. 3 is a rounded diagram explaining in detail the jump and backward control. be. The device shown in the figure is electric Wh'l1IA, and νG1 is a 4-frequency sound generator. MlfT is a circuit network, VRIB is a voice response unit, NOV is a network control device, HYB is a hybrid, four-frequency sound detector, v
oo is the speech synthesis section, OOM? In the control section, the power is turned on.

Claims (1)

[Claims] In a voice response system having a telephone and a voice response device, in response to a call from the telephone, nine sentences prepared in advance are successively prepared in a speech synthesis section in the voice response device and the telephone is answered by voice. , blocks the sentences in the cough speech synthesis unit into multiple sentence groups, adds a sentence group number and a sentence number to each sentence, and sends a sentence from the control unit in response to a sentence skip command or a sentence backward command from the phone. A voice response system characterized in that the sentence group number or sentence number of the 9th sentence is read out by ±1 and the sentence numbered 9th or the sentence is outputted.