JPH0722917Y2 - Voice conversion phone - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial application] The present invention relates to a voice conversion telephone.

[Prior Art] A telephone is formed so as to faithfully transmit a voice signal input from a transmitter to the other party in order to reliably communicate with the other party.

However, in the past, since the telephone itself was regarded as a practical one, various ideas were applied to the design and the original function of the telephone itself, but the telephone itself was not regarded as an object of play.

[Problems to be Solved by the Invention] The present invention has been made in view of such a viewpoint, and its purpose is to enjoy conversation with the other party by using a voice different from one's original voice. It is to provide a telephone with a voice conversion function capable of performing.

[Means for Solving Problems] In order to achieve the above object, the voice conversion telephone according to claim 1 is a voice conversion circuit for converting a voice signal input from a transmitter into a different voice signal, and the voice input. A switch for selecting a signal or a voice signal converted by the voice conversion circuit, and by switching this switch, the converted voice signal is replaced with the voice signal input from the transmitter. The voice conversion circuit is formed so as to be transmitted to the other party, and the voice conversion circuit converts an voice signal input from a transmitter into a digital signal, and a digital signal converted by the analog-digital conversion unit. A memory for storing data in an area designated by a write address, and converting a digital signal read from the memory into an analog audio signal A digital-to-analog converter that performs a fixed-cycle counting operation, a first counter that specifies one of a write address and a read address for the memory, and a variable-cycle counting operation. , A second counter for designating the other of the write address and the read address to the memory, and a level set value for determining the pitch level by an external operation are input to set this level set value. A level setting unit that sets a variable cycle of the second counter based on the level setting unit; and when the count cycle of the second counter set by the level setting unit is shorter than the count cycle of the first counter, Instead of the level set value input by the level setting unit, the complement of this level set value is calculated, and this calculated When the complement calculation unit that sets the count cycle of the second counter based on a number, and the count cycle of the second counter that is set based on the complement calculated by the complement calculation unit A switching unit for interchanging a write address and a read address designated by the first and second counters and inputting the read address to the memory, and changing a relative speed of the digital signal to be read from and written to the memory. This is characterized by changing the pitch of the audio signal.

The voice conversion telephone set according to claim 2 is the voice conversion telephone set according to claim 1, wherein the voice conversion circuit includes a peak detection unit that detects a peak of a level of an input voice signal, and the peak detection unit for each peak detection. It is characterized by further including a control unit for performing control to write the final address data to the position of the write address of the memory and reset the write address.

The voice conversion telephone set according to claim 3 is the voice conversion telephone set according to claim 2, wherein the peak detection unit has a constant dead zone time shorter than one cycle of the voice signal after detecting the first peak. It is characterized in that only the first of a plurality of peaks included in one cycle of an input audio signal is detected.

[Operation] In the voice conversion telephone according to the invention of claim 1, the cycle of the second counter is variably set and the cycle of the first counter is fixed by the level conversion section operated from the outside. Then, the read / write address for the memory is specified by these two counters, and the relative speed of the digital signal for reading / writing the memory is arbitrarily changed.

With this, it becomes possible to talk with the other party by an arbitrary voice different from the original voice by an operation from the outside.

When the count cycle of the second counter that is variably set becomes short, the complement of this level set value is calculated instead of the level set value input by the level setting unit, and the calculated complement The count cycle of the second counter is set, and the write address and the read address are exchanged by the switching unit.

Therefore, for example, when the read address is specified by the second counter, even if the resolution of the pitch level is reduced due to the shortened count cycle, the cycle for specifying the read address is shortened. Further, by lengthening the period for designating the write address, high resolution can be maintained.

In the voice conversion telephone according to the invention of claim 2, the write address to the memory is reset every time the peak of the voice signal input from the transmitter is detected.

As a result, almost the same level of audio signals are stored in the same storage location of the memory, so that even if the read address overtakes the write address or vice versa, the generated tremolo modulation noise is generated. Can be kept to a minimum.

Further, in the voice conversion telephone according to the invention of claim 3, immediately after the above-mentioned peak detection, there is a fixed dead zone time shorter than one cycle of the voice signal, and a plurality of peaks appear in one cycle. Even in such a case, it is possible to accurately detect the first peak and effectively reduce the tremolo modulation noise.

[Embodiment] Next, a preferred embodiment of the present invention will be described with reference to the drawings.

Embodiment FIG. 2 shows a preferred example of a voice conversion telephone to which the present invention is applied. The telephone of the embodiment is the telephone body
By operating the original function key 12 of the telephone, which is provided on the main body 10 side, and has the handset 10 and the handset 14, the telephone handset can be transmitted and received in the same manner as an ordinary telephone.

A feature of the present invention is that it enables the user to talk with the other party using a voice that is different from the one he or she normally uses, and by doing so, while enjoying the change in one's own voice. You can have a conversation with the other party.

In this embodiment, it is formed so as to talk to the other party with a voice in which the pitch of the input voice signal is converted. Therefore, the telephone of the embodiment has a changeover switch 16 for pitch conversion.
Is provided, the voice conversion function is exhibited by turning on the changeover switch 16, and the pitch conversion function indicator 18 lights up.

Further, the telephone body 10 is provided with a pitch adjusting volume 20, which is formed so that the pitch conversion level of the input audio signal can be freely adjusted.

Furthermore, the telephone of the embodiment is provided with a ping pong button 22 and a boo button 24. By pressing these buttons 22 and 24, the synthesized sounds of "ping pawn" and "boo" are mixed into a normal conversation and transmitted to the other party. It

Therefore, using the telephone of the present invention, for example, while calling between friends, while enjoying the question and answer of the quiz with a voice different from the original voice, in addition to this, the correctness of the answer " You can enjoy it by notifying the other party with synthetic sounds such as "ping pong" and "boo".

For example, if you ask a question such as "Who am I?" With a voice different from your original voice, and if the other party correctly guesses your name, press the Ping Pong button 22,
If you make a mistake, you can play by pressing the boo button 24.

FIG. 1 shows the circuit configuration of the telephone of this embodiment. The telephone of this embodiment has a telephone circuit 30 and a voice conversion circuit 32.
Have.

When a normal telephone call is made, the changeover switch 16 is turned off and the voice signal input from the transmitter 14a is directly input to the telephone circuit 30 as indicated by arrow A.

Further, when converting the voice, the changeover switch 16 is turned on, and the signal input from the handset 16 is input to the voice converting circuit 32 as shown by an arrow B. The voice conversion circuit 32 is formed so as to convert the input voice signal into a different voice and output it to the telephone circuit 30.

As such a voice conversion circuit 32, various circuits can be used as needed, and in the embodiment, a circuit for converting and outputting the pitch level of a voice signal is used. The pitch level to be converted can be arbitrarily set by operating the pitch adjusting volume 20 shown in FIG.

Since the telephone of the embodiment is formed in this way, for example, when making a normal telephone call, the changeover switch 16 may be turned off.

As a result, the voice signal input from the transmitter 14a is directly input to the telephone circuit 30 as indicated by the arrow A, and the telephone line 10
It is transmitted to the other party via 0 and is input to the handset 14b. The voice signal sent from the other party is
It is input to the handset 14b via the telephone circuit 30.

In this way, the telephone of the embodiment can have a normal conversation.

Also, when changing the pitch of the voice and making a call to the other party, the changeover switch 16 may be turned on.

As a result, the voice signal input from the transmitter 14a is input to the voice conversion circuit 32 as shown by the arrow B, converted into a desired pitch level here, and then input to the telephone circuit 30. Then, the voice signal converted from the telephone circuit 30 is transmitted to the other party via the telephone line 100 and is input to the handset 14b.

In this way, the speaker can have a conversation with the other party by converting his or her own voice into a voice with an arbitrary pitch, enjoying the voice, and enjoying the reaction of the other party. Therefore, if the speaker is, for example, a male, the pitch can be converted to a voice that is higher or lower than the usual one, and the conversation can be enjoyed while listening to that voice. You can also convert it and enjoy the reaction of the other party.

On the contrary, even when the speaker is a woman, his / her voice can be adjusted to a higher or lower level, or the voice can be converted into a male voice to enjoy the conversation.

Thus, by using the telephone of the present invention, it is possible to change the communication by telephone and enjoy the telephone itself as entertainment.

Especially when adjusting the pitch level as in the example,
Depending on the adjustment, you can freely produce different voices depending on gender, age, body type, etc., you can enjoy talking with each other while imitating the voices of popular stars such as TV, movies, etc. You can also simulate other people.

The present invention can also be used as a welfare device for language-impaired persons because it can process speech that is difficult for a language-impaired person to hear into a voice signal that is easy to hear and transmits it to the other party. Also, in the case of a woman, the prank call can be repelled by converting her voice into a male voice and speaking.

The present invention is not limited to the above embodiment,
Various modifications can be made within the scope of the present invention.

For example, although the embodiment has been described by taking as an example the one that converts the pitch level of the audio signal as the audio conversion circuit 32, the present invention is not limited to this, and in addition to this, as such an audio conversion circuit 32, a digital delay, Various conversion circuits such as a flanger, a phaser, a parametric equalizer, and a vocoder can be used, and these various conversion circuits can be used in combination.

For example, if a digital delay is used as the voice conversion circuit 32, an echo can be added to the voice signal, and as a result, the voice can be converted into a voice that speaks in a flow field, for example.

Further, if the flanger is used as the voice conversion circuit 32, the timbre of the voice signal can be converted, and for example, the voice signal can be converted into a noisy robot voice and output.

Further, by using the phasor, it is possible to convert the voice signal into a voice with a high swell (rotation effect) and output the voice.

Also, an equalizer can be used as the voice conversion circuit 32. At this time, if a parametric equalizer is used,
The change amount of the audio signal can be made larger than that of the equalizer for audio, and the audio signal can be converted and output to a sound coming from an AM radio, for example.

Specific Example of Voice Conversion Circuit Next, as the voice conversion circuit 32, a circuit configured to convert a pitch level will be taken as an example to describe a specific configuration thereof.

FIG. 3 shows a specific example of the voice conversion circuit 32 configured to convert the pitch level.

The voice conversion circuit 32 of the embodiment includes a memory 40, a read address counter 42A, a write address counter 42B, and
And a switch 44 that alternately outputs the outputs of the counters 42A and 42B and outputs them to the memory 40.

Then, the voice signal 300 input from the transmitter 14a shown in FIG. 1 is converted into a digital signal via the amplifier 46B and the A / D converter 48B and is input to the memory 40. Then, the input signal is sequentially written and stored in the memory area of the memory 40 designated by the address counter 42B.

The audio signal written in the memory 40 is sequentially read from the memory 40 in accordance with the read address designated by the address counter 42A, converted into an analog signal by the D / A conversion circuit 48A, and then the amplifier 46A. Telephone circuit through 3
It is output to 0.

FIG. 5 shows an example of a memory map of the memory 40. The address counters 42A and 42B sequentially specify addresses from "0" to "n", and the addresses are changed to "n". When it reaches, it is formed so that addresses are sequentially designated again from "0".

Therefore, when the speed at which the count values of both the write and read counters 42B and 42A are incremented is the same,
The input voice signal is output as it is at the same pitch level.

If the increment speed of the write address output from the address counter 42B is faster than the increment speed of the read address output from the address counter 42A, the input audio signal 300 is converted to a low pitch and output. Will be. This is input when the read address is incremented by 0,0,1,1,2,2, ... while the write address is sequentially incremented by 0,1,2,3,4 ,. This is because the frequency of the output audio signal is relatively lower than that of the audio signal.

On the contrary, when the increment speed of the write address is slower than the increment speed of the read address, the frequency of the output audio signal is relatively higher than that of the input audio signal, and the pitch level is high. Will be converted and output.

Such a pitch level conversion output method is called a cut and splice method. In such a circuit, if the increment cycle of each of the address counters 42A and 42B is controlled, the input voice signal 300 can be output as a voice signal having an arbitrary pitch.

Therefore, the voice conversion circuit 32 of the embodiment is provided with an oscillator 50 that oscillates and outputs a reference clock, counters 52A and 52B that count and output the reference clock, a comparator 54, and a pitch level setting circuit 56. There is.

The counter 52B divides the basic clock output from the oscillator 50 into 1/8 and outputs the divided output to the address counter 42B.

The counter 52A is formed as a 4-bit counter, counts the basic clock output from the oscillator 50, and outputs the count value to the comparator 54.

Further, the pitch level setting circuit 56 is set to 1 from “1” to “15” in accordance with the count range of the 4-bit counter 52A.
It is formed so that the set value can be adjusted in five steps, and the set value is output to the comparator 54 as a 4-bit modulation degree digital signal.

Then, the comparator 54 outputs a pulse signal to the address counter 42A and resets the count value of the 4-bit counter 52A at the same time when the count output of the 4-bit counter 52A reaches the set value of the pitch level setting circuit 56. .

Therefore, the set value of the pitch level setting circuit 56
If it is set to "8" according to the frequency division cycle of 52B, the comparator 5
From 4, pulses are output at the same cycle as the frequency division counter 52B. Therefore, in this case, the increment speeds of the count values of the address counters 42A and 42B are the same, so that the input voice signal 300 is output at the same pitch level without any change.

Further, when the set value of the pitch level setting circuit 56 is set to a value higher than "9", the cycle in which the pulse is output from the comparator 54 becomes longer than the cycle in which the count pulse is output from the frequency dividing counter 52B.

Therefore, in this case, the increment speed of the read address output from the address counter 42A becomes slower than the increment speed of the write address output from the address counter 42B, so that the audio signal is converted and output to a low pitch. .

On the contrary, when the set value of the pitch level setting circuit 56 is set to a value of "7" or less, the input audio signal is
It will be converted into a high pitch and output.

By the way, in this audio conversion circuit 30, when the set value of the audio level setting circuit 56 is set to "15", the audio signal does not reach one octave, but can be converted and output to a low tone close to it, When the value is set to "4", the sound is converted and output one octave higher.

Therefore, in this circuit, as shown in FIG. 6, when converting and outputting the pitch level to the lower one, it is possible to obtain 8 levels of modulation resolution per octave, but adjust the pitch level to the higher one. In this case, there is a problem that only four levels of modulation resolution can be obtained per octave, and the pitch level adjustment becomes extremely rough.

FIG. 4 shows a voice conversion circuit 32 that solves such a problem.
An example is shown, and the circuit of the embodiment includes an inverter 60
And a read / write control circuit 62.

Then, the inverter 60 determines whether to convert the pitch level to a higher level or a lower level, based on the modulation degree digital signal output from the pitch level setting circuit 56. Then, only when converting and outputting to the higher one, the complement of “1” of the input modulation degree digital signal is created and output to the comparator 54.

Further, the writing / reading control circuit 62 determines whether to convert the audio signal to the higher one or the lower one based on the modulation degree digital signal as described above.

Then, when converting to the lower one, as in the circuit shown in FIG. 3, switching is performed so that the address output from the address counter 42A is the read address and the address output from the address counter 42B is the write address. The device 44 and the memory 40 are controlled.

On the contrary, when converting the audio signal to the higher one, the switching unit 44 and the memory 40 are controlled so that the output of the address counter 42A is used as the write address and the output of the address counter 42B is used as the read address. . At this time, the modulation degree digital signal set and output from the pitch level setting circuit 56 is set to "1" by the inverter 60 as described above.
Is converted to the complement of and input to the comparator 54, the comparator 54
Will function as a frequency dividing counter for frequency-dividing and outputting the basic clock over a total of 7 stages of 8 to 15.

Therefore, by appropriately changing the frequency division cycle using the pitch level setting circuit 56, the count cycle of the read address output by the address counter 42B can be changed to the address counter.
7 for the count cycle of the write address output by 42A
The speed can be relatively increased over the steps, and pitch level control with high resolution can be performed even when converting an audio signal to a higher one.

By doing so, as shown in FIG. 7, the read address count value and the write address count value input to the memory 40 according to the value of the modulation degree digital signal output from the pitch level setting circuit 56. That is, the frequency division cycle, that is, the increment cycle is controlled.

Therefore, according to the present embodiment, it is possible to convert the pitch level with high modulation resolution regardless of whether the audio signal is converted to the higher one or the lower one.

Tremolo Modulation Noise Countermeasure Further, in the pitch level conversion circuit using the cut-and-splice method as in the present embodiment, the read address periodically overtakes the write address when the pitch level is converted and output high. In addition, when the pitch level is converted and output low, the write address periodically overtakes the read address.

As described above, when one address passes the other address, some tremolo-modulated sound may be generated as noise depending on the pitch conversion level setting.

In the present embodiment, such a tremolo modulation noise is reduced by using a circuit as shown in FIG.

That is, as shown in FIG. 10, the human voice signal has a waveform in which adjacent waveforms are similar to each other.

Therefore, at the same time that the peak of the input audio signal is detected, the audio signal is sequentially written from the address 0 of the memory 40, and when the next peak is detected, the data is written again from the address 0. For example, signals of the same level are written in each storage area of the memory 40,
Even if the read address has passed the write address and vice versa, the tremolo modulation noise generated can be minimized.

In order to perform such a write / read operation, the circuit shown in FIG. 8 uses the peak detection circuit 70 to detect the peak value of the audio signal based on the audio signal input via the amplifier 46B. Is input to the control circuit 72.

When the peak value is detected, the control circuit 72 inputs the final address data into the area of the write address input to the memory 40 at this time and resets the write address output from the write address counter to "0". To do.

When the final address data is read by the read address output from the read address counter, the read address is reset to "0" at the same time.

By doing so, at the same time that the peak value of the input audio signal is detected in the memory 40, the audio signal is sequentially written from the address 0 of the memory 40 and the next peak is detected. Then, data writing is performed again from the address 0. Therefore, in each storage area of the memory 40,
A signal of similar level will be written. Further, the data thus written is sequentially read in accordance with the read address.

Note that such an audio signal may have a plurality of peaks locally as shown in FIG. 9, for example.

Therefore, the peak detection circuit 70 of the present embodiment sets a dead zone time for a certain period after detecting the first peak, and determines that all the peaks generated within this dead zone time are invalid.

By doing this, even if there are multiple local peaks,
Without being affected by this, the peak of one waveform to the peak of the next waveform can be accurately detected, and tremolo modulation noise can be effectively reduced.

[Effects of the Invention] As described above, according to the present invention, the voice signal input from the transmitter can be converted into a different voice and transmitted to the other party, so that the speaker can enjoy the converted own voice. While enjoying the reaction of the other party, you can enjoy the call with a different feeling than usual.

In particular, if it is configured to convert the pitch of the input audio signal and send it to the other party, you can set your voice high or low as necessary, listen to the converted voice yourself, and respond to the other party's reaction. You can enjoy the conversation while listening.

Further, if the pitch level to be converted is set low, the voice of a woman is converted into the voice of a man and is output, so that it can be used for repelling a prank call.

Further, according to the present invention, when the period for designating the write address or the read address to the memory is shortened and the resolution of the pitch level to be converted is lowered, the complement of the level setting value is used and the memory is used. By replacing the read / write address with respect to, it is possible to maintain high resolution of the audio level.

Further, when the peak of the voice signal input from the transmitter is detected, the waveforms of the voice signal are stored from the same address position in the memory in the same address order, so that almost the same level is stored in the same storage location in the memory. Since the audio signal is stored, the tremolo modulation noise generated can be minimized even when the read address overtakes the write address or vice versa.

Further, by providing a fixed dead zone time shorter than one cycle of the audio signal immediately after the peak detection, the first peak can be accurately detected even when a plurality of peaks appear in one cycle. Tremolo modulation noise can be effectively reduced.

[Brief description of drawings]

FIG. 1 is a telephone circuit diagram showing a preferred example of a voice conversion telephone according to the present invention, FIG. 2 is an external view of a telephone according to this embodiment, and FIGS. 3 and 4 are used for the above embodiment. A concrete block circuit diagram of the pitch conversion circuit, FIG. 5 is an explanatory diagram of a memory used in FIGS. 3 and 4, and FIGS. 6 and 7 are circuits of the circuits shown in FIGS. Operation explanatory diagram, FIG. 8 is a circuit explanatory diagram for reducing the tremolo modulation noise used in the circuits of FIGS. 3 and 4, and FIGS. 9 and 10 are explanatory diagrams of the circuit operation shown in FIG. is there. 14 …… Handset 16 …… Changeover switch 20 …… Pitch adjustment volume 30 …… Telephone circuit 32 …… Voice conversion circuit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Hideyo Oguri Inventor Hideyo Oguri 2-8-5 Tamagawa, Ota-ku, Tokyo NAMCO INC. (56) References JP 62-34441 (JP, A) JP 59- 94953 (JP, A) JP-A-57-114198 (JP, A) JP-A-59-110013 (JP, A)

Claims (3)

[Scope of utility model registration request] 1. A voice conversion circuit for converting a voice signal input from a transmitter into a different voice signal, and a switch for selecting the input voice signal or a voice signal converted by the voice conversion circuit. By switching this switch, it is formed so as to transmit the converted voice signal instead of the voice signal input from the transmitter to the other party. An analog-digital converter that converts an input audio signal into a digital signal, a memory that stores the digital signal converted by the analog-digital converter in an area designated by a write address, and a memory that is read from the memory A digital-to-analog converter that converts a digital signal into an analog audio signal, and a fixed-cycle count operation A first counter for designating one of a write address and a read address for the memory, and a second counter for designating a write address or a read address for the memory by performing a variable cycle count operation. By inputting a level set value for determining the pitch level by an external operation, the level setting for setting the variable cycle of the second counter based on this level set value. And a level setting section input by the level setting section when the count cycle of the second counter set by the level setting section becomes shorter than the count cycle of the first counter. Complement calculation of calculating the complement of the set value and setting the count cycle of the second counter based on the calculated complement And a write address and a read address designated by the first and second counters when the count cycle of the second counter is set based on the complement calculated by the complement calculator. A voice conversion telephone, comprising: a switching unit which replaces and inputs to the memory, and changes the pitch of the voice signal by changing the relative speed of the digital signal read from and written to the memory. 2. The peak conversion unit according to claim 1, wherein the audio conversion circuit detects a peak of a level of an input audio signal, and a peak detection unit detects a peak at a write address position of the memory each time the peak is detected. A voice conversion telephone characterized by further comprising a control unit for performing control of writing address data and resetting a write address. 3. The peak detection section according to claim 2, wherein the peak detection section has a fixed dead zone time shorter than one cycle of the audio signal after detecting the first peak, and A voice conversion telephone characterized by detecting only the first of a plurality of peaks included in a cycle.