JPS6027269A - Detection for the other-party signal at audio response signal transmission time - Google Patents

Abstract

PURPOSE:To detect the other party request signal without malfunction by providing a silent section between phrases or the like of a voice transmitted from own device or putting restrictions to the transmission spectrum of the voice. CONSTITUTION:Silent sections P1-P3, which do not make the voice unnatural, are provided between phrases of the voice transmitted from own device, and detection windows W1-W3 are provided in these silent sections P1-P3 to detect the other party signal. Otherwise, restrictions are put on the transmission spectrum of the voice, and a limit is provided in a band lower than this controlled spectrum fc, and the band width of this spectrum fL-fc is used as a detection window, and the frequency of the other party request signal is set within this window to detect the other party signal.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention is applicable to a facsimile IJ absent reception device, etc.
The present invention relates to a means for detecting a signal without error when a request signal for forcibly interrupting the response, switching to a telephone, or switching to a facsimile is received from the other party during a voice response.

[Prior art]

In a facsimile missed receiving device, etc., when a call is received from the other party, the output of voice is automatically started by a ringing signal such as a bell. Normally, once the voice output starts, it continues for about 20 seconds, which is a waste of time for a caller who knows the content. In such a case, the person calling is the fax machine or telephone at hand.

Interrupt the audio output of the other party's absence receiving device by pressing a button or voice, connect the line to a facsimile, and start facsimile communication; or request a telephone conversation. In such a case, a voice response device such as an absentee receiving device must detect these request signals sent from the other party while outputting the voice itself.

FIG. 1 shows a conventionally used means for realizing this function. In FIG. 1, 1 is a line such as a telephone network to which a voice response device (1 not shown) is connected. On this line, voice is sent from the voice response device, and at the same time a request signal such as an interruption is output from the other party. 2 is a hybrid circuit that combines the voice signal output from the voice response device and the signal sent from the other party. It has a function of separating signals. 3. A voice generation circuit has a function of generating voice stored in a tape recorder, a voice synthesis circuit, etc., and its output is sent to line 1 via hybrid circuit 2. 4. 5 is a control circuit that starts and stops voice generation, detects a request signal from the other party, controls the overall operation, etc. 5 is a filter that is sent from the other party (
It has a bandpass characteristic that matches the frequency of the request signal from the other party, and an output here indicates that there is a request signal from the other party.

A request signal from the other party enters this filter 5 via the hybrid circuit 2. The problem with the conventional configuration as described above is that the hybrid circuit 2 is not ideal and the audio output signal that it sends leaks into the filter 5 side. It would be expensive to make this hybrid circuit 2 ideal to prevent leakage. Therefore, in order to construct the device economically, this leakage must be allowed to some extent. If there is voice leakage into the filter 5, the voice has a spectrum that includes the bandpass band of the filter 5, so a signal is generated at the output of the filter 5 by the voice transmitted by the other party, as if from the other party. Whenever there is a request signal (the signal is input to the control device 4, there is a problem that the control device 4 makes a false detection).

[Purpose of the invention]

This invention applies to the aforementioned audio signals that you send out.

This is intended to prevent malfunctions due to

[Summary of the invention]

One of the aspects of this invention is to provide a silent section (no signal section of the transmitted signal) to the extent that the phrases of the speech you transmit and the opening sounds of the phrases do not sound unnatural, and to prevent the signal from the other party from coming in during this silent section. The other method is to set a limit on the audio transmission spectrum and set the frequency of the request signal from the other party in this restricted/limited area to detect this. That is.゛Hereinafter, the present invention will be described in detail with reference to the embodiments shown in the drawings.

[Embodiments of the invention]

FIG. 2 shows a diagram of the principle of this invention. The m12 diagram shows the relationship between the passage of time and signals. Figure 2 (a)
indicates the output state of the voice output from the voice response device. Normally, speech is composed of a series of phrases, and even if there are some pauses (silent intervals) between phrases, the words do not seem unnatural. In this figure, Fl, F'! ! ...Han indicates a phrase to be uttered, and 'I*'2 t'3 indicates a silent section. During this silent period, no signal is output from your side to the line, so if there is a signal during this period, it is a signal.

It is a signal sent from the hands.

FIG. 2(b) shows an example of a request signal sent from the other party, and this signal may be a single tone, a complex tone such as a button signal, or a voice.

FIG. 2(C) shows a detection window for detecting the other party's signal during the rest interval, and the operation for detecting the other party's signal is performed during this interval W, , W, , W. 2 shows the detection result VV.

This indicates that output is occurring because there is a signal from the other party in the section marked by and crane.

FIGS. 3 and 4 are examples of the configuration of an apparatus based on the above-mentioned principle. Figure 3 shows an example in which a voice synthesis circuit or other means that is relatively easy to control the generation and stop of phrases is used as the voice generation circuit, and Figure 4 shows an example in which a voice generation circuit such as a tape recorder is used once the voice is started. This is the case when using a device that is difficult to stop and restart once it is turned on. 8 in Figure 3
, 2, 5.4 are equivalent to the constituent members (circuits) shown in FIG. When the circuit 4 instructs the voice generation circuit 3 to pronounce a phrase and start the voice generation via the control line 1o, the voice generation circuit 6 starts generating the phrase.The generated voice is transmitted through the hybrid circuit 2. It is output as an audio signal to the line 1. When the utterance of one phrase is finished, the audio signal is output from the audio generation circuit 3 to the control circuit 4.
The completion of the process is notified to the computer via the control line 11. The control circuit 4 temporarily suspends the generation of the next phrase, and the other party signal detection circuit 6
The gate circuit 7 is opened to input the signal from the gate. Signal 8 is a signal that opens gate circuit 7, and its output is input to control circuit 4 by signal 9. The control circuit 4 determines whether there is a signal from the other party during this period, and if necessary, instructs the voice generation circuit 3 to utter the next phrase, and if necessary, stops the utterance and continues the other party's signal. You can take other measures such as receiving the same. Here, the other party signal detection circuit 6 is composed of a relatively inexpensive device such as a filter or a period counter when the expected other party request signal is a single tone, and when it is a complex wave such as a pushbutton signal from a telephone etc. A dedicated detection circuit is used, and if it is a voice, it is configured with a voice identification circuit.

FIG. 4 shows an example in which a method is used to accurately start and stop the sound generation of a tape recorder or the like in the sound generation circuit 6.The control circuit 4 is connected to the sound generation circuit 6 via a control line 10. It is assumed that the voice generation circuit 6 has recorded voice with an appropriate pause period between the voice phrases.The output of the voice generation circuit 3 is sent to the hybrid circuit 2. The level detection circuit 12 generates a signal at the output 15 and controls it when the output of the signal 6 reaches a small enough value to detect the other party's signal. It is assumed that the control circuit 4 contacts the other party signal detection circuit 6 by this signal.
It is assumed that the output of the circuit 7 is taken in through the gate circuit 7.

In addition to the output of the other party signal circuit 6, the output 13 of the level detection circuit 12 is input to the gate circuit 7, and the output 9 receives the other party signal detection circuit 6 only when the output of the audio generation circuit 3 is silent or sufficient/JX. A detection output appears. By doing so, the control circuit 4 can prevent malfunction (it can detect the other party's request signal). Although the hybrid circuit 2 is used in FIGS. 3 and 4, this is not necessarily a necessary configuration (other Although the gate circuit 7 is inserted after the other party signal detection circuit 6, it may also be inserted before the other party signal detection circuit 6.

FIG. 5 shows a principle diagram of another example of the present invention. The principle shown in FIG. 2 is a method of providing a detection window for the other party's request signal in the time domain, but the fifth example is a method of providing a detection window in the frequency domain. Commercial audio has a frequency bandwidth of several tens of Hz to kHz, but when it is sent to a line, this bandwidth is limited. For example, subscriber telephone lines are limited to 3008Z to 3.41G (z). In Fig. 5, (a) shows the audio transmission bandwidth, which is the band from the lowest frequency fL to the highest frequency jtt that can pass through the line. Therefore, the voice output of voice response device f is considered to have a voice spectrum distributed over almost the entire band.It is difficult to detect only the other party's request signal from such a spectrum distribution. This is difficult unless the hybrid circuit is ideal.

The present invention attempts to detect the signal requested by the other party by limiting the transmission spectrum and setting the frequency of the signal requested by the other party within the limited spectrum.

This restriction is possible if some degree of deterioration of the transmitted audio is recognized.

It is also possible to set a limited band and bandwidth that do not feel unnatural to the human ear. Figure 5 (For example, in an example where a limit is placed on the low frequency range below fc, the diagonal line between fc and fL is the part where the transmitted voice is cut, and it does not look too unnatural with a female voice etc.. This fc Detection is possible by using the bandwidth from fr to fr as a detection window and setting the frequency of the other party's requested signal within this window.Furthermore, even if you do not intentionally set a limit, it may be possible for some voices to not have a spectrum in a specific band. It is also possible to provide a detection window in such a band. Fig. 5 (c) shows an example in which a detection window is provided between fr and fc, and 14 indicates this range, within which the other party's signal can be detected. The frequency of jp is set as .jp does not have to be a single tone (it can be a complex wave) as long as it is within this band.

Further, depending on the capability of the voice identification circuit and the bandwidth of the detection window, it may be voice.

Figure 6 shows an example of the configuration of a device based on this principle.1.2
, 3, 4, 6, 9, and 10 are structural members equivalent to those in the examples shown in FIGS. 1, 3, and 4. Reference numeral 15 denotes a bandwidth limiting circuit such as a filter for limiting the frequency band width of the output of the voice generating circuit 3, and it is assumed that the other party's request signal' falls within this limited range.

In this way, signals in this band will not leak from your own voice generating circuit 3 to the other party's signal detection circuit 6, and it will be easier if the other party's signal detection circuit 6 has the ability to detect only signals in this band. can detect the other party's signal. In the example of the 5th inch, the detection window was set at only one location, but it may be provided at multiple locations. In the example of FIG. 6, if there is a trough in the spectrum in the output of the voice generating circuit 3, the other party (!1) detection circuit 6 may be omitted and a detection window in this spectral region may be used.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to easily detect the other party's request signal without malfunctioning while transmitting the voice using the voice response device or the like.

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram of the main parts of a conventional absentee reception device, Fig. 2 is an explanatory diagram of the principle of the present invention, and Figs. FIG. 5 is a detailed explanatory diagram of the present invention, and FIG. 6 is a configuration diagram showing an embodiment thereof. 1...Line, 2...Hybrid circuit, 3. ...Sound generation circuit, 4.Control circuit, 5.6.Destination signal detection detection circuit, 12.Level detection circuit, 15.Bandwidth restriction circuit.Representative Patent Attorney Akio Takahashi Rod /l 2nd area Ft 7) r F2, Fzra pull-'-i, t? r
- 111 (+ 'l 11 11 °7" + 23 ・ 11-1 1 1.1 1111 V'Hz V31- 11 11", the power output 11 "H! !! Figure 3 4 Illusion

Claims (2)

[Claims] (1) Eight devices that automatically respond with voice through a line are characterized in that when detecting a signal from the other party, a silent period is provided for the voice generated by the device, and the signal of the other party is detected during this silent period. How to detect the other party's signal when sending a voice response signal. (2) When detecting a signal from the other party with a device that automatically responds with voice through a line - A no-signal section is created in the frequency spectrum of the voice generated by the device, and the spectrum of the other party's signal is detected within this no-signal spectrum. A method for detecting a signal from the other party when transmitting a voice response signal, characterized in that the signal from the other party is detected by setting .