JPS6239870B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] When transmitting a communication path signal in which voice and signal tones such as busy tone/ringing tone are mixed by a general voice analysis and synthesis method, the signal tone is waveform-shaped and then A signal transmission system is disclosed in which analysis and synthesis can be performed using the amount of correlation of time fluctuations in frequency components by generating harmonic components.

[Industrial application field]

The present invention relates to a signal tone transmission method that enables the transmission of signal tones such as ringing tones with a simple configuration in a voice transmission method using a voice analysis and synthesis method.

With the recent development of digital technology, a speech analysis and synthesis method has come to be used as a method for transmitting four audio lines over a conventional leased line (D-1 standard). This method uses the amount of correlation between frequency components of audio. On the other hand, signal tones such as busy tone/ring tone are basically sinusoidal waves, which consist of a single repeating frequency, and there is no time variation in frequency components. Therefore, it is impossible to transmit to the other party using the amount of correlation.

[Conventional technology]

A method of speech analysis on the transmitting side and a method of speech synthesis on the receiving side will be explained with reference to the drawings.

FIG. 4 is a block diagram showing the main parts of the speech analyzer, and FIG. 5 is a block diagram showing the main parts of the speech synthesizer.

In FIG.
is converted into a digital signal.

The digitized signal is passed through a voice analysis filter 6.
2, and outputs a parameter output x that indicates the characteristics of the voice through filtering based on correlation calculation of time fluctuations in frequency components, and a residual signal output y from which the characteristic components extracted from the signal are removed. be done. The fundamental frequency and residual signal power of the residual signal output y are analyzed by a sound source analysis section 63, and the result is output as a sound source information output z.

The audio feature parameter x and the sound source information z are multiplexed by a multiplexer 64 and output.

In FIG. 5, the input signal to the speech synthesizer 8 is processed by the separation unit 8 into the speech feature parameter x' and the sound source information.
It is separated into z′.

The sound source information z' is input to the sound source generating section 83, and a sound source pulse synchronized by the fundamental frequency and having a magnitude according to the signal power is generated, and is input to the voice synthesis filter 82 together with the sound source characteristic parameter x'. The signal synthesized by the voice synthesis filter 82 is sent to the D/A converter 84
is converted into an analog signal and output.

When a sine wave is input to the voice analysis filter 62 of the voice analyzer 6, the residual signal output after the correlation calculation cannot be obtained, and the residual signal power becomes zero as a result of the sound source analysis, so the voice on the receiving side It becomes impossible to combine the signals in the combiner 8.

That is, a signal sound such as a busy tone/ringing tone, which basically consists of a sine wave signal, has only a single frequency component, so that the voice analysis filter 62 cannot obtain a correlation calculation output.

Therefore, when transmitting a channel signal containing a mixture of voice and signal tone using the signal analysis and synthesis method described above, a circuit for detecting and identifying this signal tone is used to separate it from the voice, and the signal tone is separated from the voice. It was meant to be transmitted as information.

However, in such conventional technology, the above-mentioned detection and identification circuit and a processing circuit for transmitting only the signal sound as information in a different format are required, making the device complicated and expensive. do.

The purpose of the present invention is to eliminate the above-mentioned drawbacks by pre-processing the signal sound using a simple circuit.
The object of the present invention is to provide a signal tone transmission system that allows signal tone and voice to be handled as data equivalent to each other.

[Means for solving problems]

The principle of the signal tone transmission system according to the present invention will be explained with reference to FIG.

The communication path 2 from the originating exchange 1 is branched, and one side is input to the waveform shaping circuit 4. Switcher 5
uses the output of the SS line (a line for notifying the presence or absence of a signal tone) 3 from the originating exchange 1 to input the input to the voice analyzer 6 to the output of 4 from the waveform shaping circuit or the above call. Switch to the signal from Route 2.

[Effect]

According to the present invention, the signal tone in the communication path signal is shaped into a rectangular wave or a triangular wave in the waveform shaping circuit 4 and then input into the voice analyzer 6.

Therefore, even if the signal sound consists of only a single frequency component, harmonic components are generated by the above-mentioned waveform shaping, making it possible to perform analysis using the amount of correlation of time fluctuations in frequency components. .

〔Example〕

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

In the figure, 1 is the originating exchange, 2 and 9 are the communication paths, and 3
Reference numeral 4 indicates an SS line which is a transmission line for the S and R signal systems, 4 a waveform shaping circuit, 5 a switch, 6 a speech analyzer, 7 a transmission path, and 8 a speech synthesizer.

Figure 2 A to D are signal waveforms and shaped waveforms.
In the example, A is a waveform indicating a ringing tone, B is a shaped waveform of waveform A, C is a waveform indicating a busy tone, and D is a shaped waveform of waveform C. A in A and C of this figure is a threshold value, and when this threshold is exceeded, a rectangular waveform is generated.

The SS line 3 is a signal tone detection line for informing the called party that a signal tone such as a busy tone/ringing tone is being sent through the communication path 2. Therefore, the signal tone on the communication path 2 is detected using the output of the SS line 3, and the switching device 5 is operated to transmit the waveform-shaped signal tone from the waveform shaping circuit 4 to the voice analyzer 6. It becomes possible to input the information into the . For example, in the case of the waveform of the ringing tone shown in FIG. 2A, it consists only of signals with a substantially constant frequency.
By shaping the waveform into the waveform (rectangular wave) shown in , it is possible to generate harmonic components and perform analysis using the amount of correlation in the same way as for speech.

Also, in the case of the busy tone waveform shown in Figure 2C,
Although there is a correlation due to fluctuations in the amplitude, the frequency components are approximately constant, and the frequency components cannot be analyzed as they are. Therefore,
Waveform shaping is performed in the same manner as above to obtain the shaped waveform (rectangular wave) shown in FIG. 2D.

The signal tone and voice shaped into a rectangular wave are analyzed and encoded in a voice analyzer 6, and then transferred to the receiving side via a transmission line 7. On the receiving side, voice synthesizer 8
After decoding the received signal, the received signal is transferred to the destination exchange via the communication path 9.

As described above, the output waveform of the waveform shaping circuit 4 is a rectangular wave or a triangular wave, and contains higher harmonic components compared to FIGS. 2A and 2C. Therefore, the combined output of the communication path 9 also contains harmonic components. Although there is no problem with the caller recognizing the signal tone even if it contains harmonic components, it is preferable to use a sine wave due to issues such as timbre.

Figure 3 is a block diagram of a receiving section with a low-pass filter to achieve this purpose.
is the same as the one with the same number in Figure 1, 10 is a reduction filter (fc = about 432Hz), 11 is a switch,
By controlling the SR line 14, which is the reception line of the S, R signal system, the voice synthesizer 8 and the communication path 13 are activated during a call (when receiving voice), and the low-pass filter is activated when there is no call (when receiving a signal tone). The output of the device 10 is connected to the communication line 13. 1
2 indicates the originating exchange.

That is, by receiving the portion of the output of the communication path 9 that corresponds to the signal tone through the low-pass filter 10 and suppressing harmonic components, it is possible to obtain an output with a sound quality close to that of the original signal tone. can.

〔Effect of the invention〕

According to the present invention, in a voice transmission method using a voice analysis and synthesis method, signal tones can be detected using a simple pre-circuit, and existing control signal detection lines (SS line, SR line).
Since it is possible to convert the waveform into a waveform that can be directly transmitted by using the signal, there is no need for a detection identification circuit and information processing especially for signal sound detection, and it can be constructed at low cost, which is very effective.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention;
Figures A to D are examples of signal waveforms and shaped waveforms, where A is a calling waveform, B is a shaped waveform of A, C is a busy tone waveform, D is a shaped waveform of C, and Figure 3 shows a preferred receiving section. FIG. 4 is a block diagram showing the main parts of the speech analyzer, and FIG. 5 is a block diagram showing the main parts of the speech synthesizer. In Figure 1, 1 is a transmitting exchange, 2 and 9 are communication lines, 3 is an SS line, 4 is a waveform shaping circuit, 5 is a switch, 6 is a speech analyzer, 7 is a transmission line, and 8 is a speech synthesizer. show.

Claims (1)

[Scope of Claims] 1. In a voice transmission system based on voice analysis and synthesis that utilizes the amount of correlation of time fluctuations in frequency components, a waveform that generates harmonic components by performing waveform shaping on a signal tone included in a communication path signal. A signal tone transmission system comprising: a shaping circuit 4; and a switching device 5 for inputting the signal tone output whose waveform has been shaped by the signal tone detection line 3 output from the exchange 1 to a voice analyzer 6.