JPH04167650A - Audio response recognizing device - Google Patents

Abstract

PURPOSE:To prevent an incongruous signal from being recorded on a voice mail part by automatically stopping a recording operation after the lapse of a silent section when no sound is detected at a sound detecting part. CONSTITUTION:The sound detecting part 11 provided with a timer function detects the presence/absence of sound from a user in a state where massage input to the voice mail part 5 is disabled. A start point detection signal P is outputted to a control part, and furthermore, when the sound is interrupted i.e., when a timer is operated from an end point and the silent section T of constant time elapses, a stopping signal R is outputted to the control part. When the input of a massage to the voice mail part by the user is completed and vocalization is stopped, the sound detecting part detects the interruption of the sound of the user, and the control part stops the operation of the voice mail part after the lapse of constant time from a time when the sound is interrupted, and only the sound of the user is recorded, and an incongruous signal tone can be prevented from being recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a voice response recognition circuit having a voice mail function using a telephone as a terminal.

[Prior Art] Generally, as this type of voice response recognition circuit, the one shown in FIG. 2 is known. To explain the overall configuration of this voice response recognition circuit, reference numeral 1 denotes a control section, which controls various sections to be described later. 2 is a hard disk drive device (r
HDDJ) records voice messages from users as digital signals. Reference numeral 3 denotes a speech synthesis section, which issues various synthesized speech messages to the user under the control of the control section 1. Reference numeral 4 denotes a voice recognition unit, which outputs an instruction signal corresponding to the instruction voice to the control unit 1 in response to an instruction voice from the user (instruction voice in response to the synthesized voice message from the voice synthesis unit 3). Reference numeral 5 denotes a voice mail section. This voice mail section 5 is configured with a buffer section, and temporarily stores and holds voice messages from the user, converts them into digital signals, and sends them to the HDD 2 via the control section 1. It's like recording. 6 is a PB signal detection unit that detects a PB double signal (button signal) when a user is using a button telephone, and outputs it to the control unit 1; 7
is a network control unit (rNcU)
), it performs line connection control, that is, incoming call detection and outgoing call control, and controls the line 8 connection.

The voice response recognition circuit configured as described above operates as follows. Note that here, a case where a user records a message in the voice response recognition circuit will be explained as an example.

When the NCU 7 detects a calling signal from the user's telephone, it hooks off and connects the line 8.

Next, the control section 1 controls the speech synthesis section 3 to issue a message to the user using synthesized speech.

When a user is using a button telephone, he or she responds to a message from the voice synthesis section 3 by pressing the button. The PB signal detection unit 6 detects the PB double signal and outputs it to the control unit 1 as a command from the user. Then, by repeating the exchange of the synthesized voice message from the voice synthesis unit 3 and the PB double signal from the user,
A mailbox number is specified. Thereafter, the voice synthesis section 3 utters to the user that the message can be sent manually, and the voice mail section 5 is activated.

Next, when the user finishes manually inputting a message to the voice mail section 5, a termination operation is performed. That is, for example, the 0. Press # etc. to issue a signal to end. Upon detecting this, the PB signal detection section 6 issues an end signal to the control section 1, which causes the voice mail section 5 to stop recording, and at the same time, the voice mail section 5 sends the message content converted into a digital signal. HD via control unit 1
Record in D2. Furthermore, by controlling NCU7, line 8
disconnect.

In the case of a dial type telephone, instructions are given by voice instead of the PB double signal, the voice signal is detected by the voice recognition section 4, and various instruction signals are outputted from the voice recognition section 4 to the control section 1.

Furthermore, in the case of this dial-type telephone, there is no end operation even when the message input is finished; for example, 3
The voice mail section 5 is operated for a minute and then automatically stopped.

[Problems to be Solved by the Invention] However, in the conventional voice response recognition device, when the button is pressed, a PB double signal is issued to issue an end instruction operation when the button is pressed.
There is a problem in that the message is recorded together with the PB double signal for ending the message.

Furthermore, the user needs to output a PB-multiple signal in order to instruct termination, which places an extra burden on the user and causes erroneous operations (erroneous operations for termination).

Furthermore, in the case of a rotary telephone, the message is stored in the voice mail section 5 by operating the voice mail section 5 for only 3 minutes, so even if the user finishes within 3 minutes, the message is stored in the voice mail section 5. There is a problem in that the voice mail unit 5 continues recording operation for only 3 minutes and records the busy tone as it is.

The present invention has been made in view of the above problems, and it is an object of the present invention to provide a voice response recognition circuit that prevents recording of unpleasant signals and is easy to use.

[Means for Solving the Problems] In order to achieve the above object, the present invention is applied to a voice response recognition device having a voice mail section for recording voice messages from a user.

Such a voice response recognition circuit includes a voice detection section that detects a utterance state, and a control section that stops the operation of the voice mail section after a predetermined period of time has elapsed after a voiced period has elapsed in the voice detection section. It is characterized by

[Function] With the above configuration, when the user finishes inputting a message into the voicemail section and stops speaking (voice), the voice detection section detects that the user's voice is cut off, and the control section To stop the operation of a voice mail part after a certain period of time has elapsed from the time when the voice trail ends, to record only the user's voice, and to prevent unpleasant signal sounds from being recorded.

[Embodiment] A preferred embodiment of the present invention will be described below with reference to FIGS. 1, 3, and 4. The overall configuration of the voice response recognition circuit of this embodiment is almost the same as the conventional voice response recognition circuit described above, so the same members are denoted by the same reference numerals.
The explanation will be omitted.

Reference numeral 11 in the figure denotes a voice detection section having a timer function, which detects the presence or absence of voice from the user in a state where a message can be sent manually to the voice mail section 5.

Then, when the user's voice is detected, that is, as shown in FIG.
), when the start of the input audio wave (sound section) is detected, the
The starting edge detection signal P shown in FIG. 3(b) is output to the control unit 1. Roughly speaking, when the sound ends, that is, when the end of the input sound wave is detected, a timer is activated from this end and outputs a stop signal R to the control unit 1 after a certain period of silence T has elapsed. . Here, the silent interval T is used when a typical user records a message, after recording a message, checking whether or not everything has been recorded, and thinking about whether there is anything else to record. This is a sufficient time, and is usually set to about 3 to 58.

The control section 12 is provided with a processing function for recording a message as shown in FIG.

Next, the operation of the voice response recognition circuit of this embodiment will be explained.

The operation of the voice mail section 5 up to box number specification is the same as that of the conventional voice response recognition circuit described above, so only the characteristic features of this embodiment will be described in the fourth section.
The explanation will be centered on the flowchart in the figure. Further, the processing in this flowchart is performed after the box number is specified.

After the hook number has been designated n, an input enable signal indicating that the user is ready to input a message is output to the speech synthesis section 3 (Sl). If the human power enable signal is output, the voice detection unit 11 is activated (S2) and monitored until the voice from the user is detected (S3). When the voice is detected, the voice detection unit 11 outputs a start detection signal P, and based on this start detection signal P, the voice mail unit 5
(S4), and the message from the user is stored in the buffer area of the voice mail section 5. Next, it is monitored whether the stop signal R is output from the voice detection section 11 (S5). On the other hand, the voice detection unit 11 monitors whether the user's voice has stopped, and if the voice has stopped, a timer is started to count from that point on, and a stop signal R is issued when the silent interval T has elapsed. Output.

When the control section 1 detects this stop signal R, it stops the recording operation in the buffer area of the voice mail section 5 (S6).

The subsequent operation is similar to that of the conventional voice response recognition circuit described above.

As described above, the voice detection unit 11 is provided to detect the presence or absence of voice from the user while recording a message, and when the voice is detected, the voice mail unit 5 starts the recording operation, and when the voice disappears, the silent section Even though the recording operation is automatically stopped after T has elapsed, only the user's message is input to the voice mail section 5, and the PB double signal for termination, which is an illegal signal, is recorded. disappears.

In addition, the user does not need to perform an operation to finish recording the message, so there is no extra burden on the user, erroneous operations are reliably prevented, and an easy-to-use voice response recognition circuit can be provided.

In the above embodiment, the voice detection section 11 was provided with a timer function and the voice detection section 11 outputted the stop signal R, but the voice detection section 11 outputs the end detection signal Q at the part where the voice ends. After receiving the termination detection signal Q, the control unit 1 outputs the end detection signal Q to the control unit 1, and after the silent period T has elapsed, the control unit 1 outputs the end detection signal Q to the voice mail unit 5.
The recording operation in the buffer area may be stopped.

[Effects of the Invention] As detailed above, according to the voice response recognition circuit of the present invention, the following effects can be achieved.

When the voice detection section detects that there is no voice, the recording operation is automatically stopped after a silent period has elapsed, which prevents unpleasant signals from being recorded in the voicemail section.
Only the user's message can be entered.

In addition, the user does not need to perform an operation to finish recording the message, so there is no extra burden on the user, erroneous operations are reliably prevented, and an easy-to-use voice response recognition circuit can be provided.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing the overall configuration of a voice response recognition circuit according to the present invention, Figure 2 is a block diagram showing the overall configuration of a conventional voice response recognition circuit, and Figure 3 shows the user's voice waveform and its FIG. 4 is a time chart showing the signals output by the voice detection section, and FIG. 4 is a flow chart showing the processing of the control section. 5...Voice mail section, 11...Voice detection section, 12
...control section. Floor diagram 1 showing the voice response recognition device according to the present invention
Figure 2 shows the input voice waveform and the signal output from the voice detection section. 4L
Figure 3 Audio 2 Self 1ψ Processing Flowchart 1-) Figure 4

Claims (1)

[Scope of Claim] A voice response recognition device having a voicemail section for recording a voice message from a user, including a voice detection section for detecting a utterance state; and a control section that stops the operation of the voice mail section after the voice mail section has been activated.