JPS5829105A - Reproducing device - Google Patents

Abstract

PURPOSE:To make reproducing sound clear and to close the sound to natural sound, by making a clock pulse frequency at readout constant and changing a clock pulse frequency at write and a reproducing speed of a recording medium interlockingly. CONSTITUTION:A reproducing signal from reproducing head 6 is sampled with switched 15 and 16 and stored 17 and 18. On the other hand, a signal 25a counting 25 a photoelectric conversion signal 13a of a rotating speed of a running motor 7 of a tape 1 and a signal 26a counting 26 an output signal of an oscillator 24 are inputted to an AND circuit 27 and an FF28 to output a signal 28a. Switches 15, 16 and switches 30 and 31 to change over the output signal of the oscillator 24 and that of a gate circuit 29, are controlled with the signal 28a, and a clock pulse from generators 32 and 33 is applied to the memories 17 and 18 and signals sampled 15 and 16 are minutely split. The count number of counters 25 and 26 and the storage capacity (number of elements) of the memories 17 and 18 are set to make the coincide.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a playback device that can adjust playback time without changing the pitch of an audio signal or the like.

When playing audio signals with a general audio playback device,
The playback is performed in the same time relationship as when it was recorded. However, with the axillary playback device, the playback time is useful when you want to record the content slowly, such as when shorthand while listening to the reproduced voice, or conversely, when you want to record the voice at a high speed when the voice is spoken in a gentle manner. If the recording time is different from the recording time, the pitch of the sound will change, resulting in a loss of clarity, making it impossible to satisfy the above requirements. A possible rounding method to solve the ζO problem is to expand or compress only the playback time without changing the pitch of the sound during recording. One of them is ``Sampling method using an 11-channel head,'' and the other one is ``Audio signal O-pitch conversion device'' which the applicant has previously published in special feature @45-1285.
The present invention belongs to the latter method, as presented in ``a method of sampling a reproduced signal, dividing it into smaller parts, writing them in a storage means, and reading them out in a different time relationship than when writing''. The purpose is to provide playback devices. That is, a playback device that makes the time width of the signal that is finally expanded or compressed constant and changes the sampling time depending on a signal related to the playback speed; -41 a playback means that can change the playback speed; signal generating means for generating a signal associated with the playback speed;
sampling means for sampling the signal reproduced by the reproduction means according to the signal generated by the signal generation means; storage means for storing good signals sampled by the templating means; and storage means for storing good signals sampled by the sampling means; signal 1
It is an object of the present invention to provide a reproducing device having a subdivision means for dividing the signal into subdivisions and an output means for outputting the signal subdivided by the subdivision means.

First, the principle of the playback device according to the method to which the present invention belongs is that the playback signal is sampled, it is further divided and written into a storage means, and when it is read out, it is read out in a different time relationship from the time of writing. It will be explained that only the playback time can be expanded or compressed without changing the pitch of the sound during recording.

Now consider the case where a signal of a constant frequency f is recorded and reproduced. If the wrinkle signal is played back at the same speed as when it was recorded and is round, the 1st m
A reproduced signal with a frequency f such as 1 (up to) K is obtained.

Then, as shown in the first WJ(b)K of the sampling signal Ka sampled from the reproduced signal at a constant time T, a sine wave exists for Tf cycles during the wrinkle constant time 10. Also, if the playback time is made to be twice the recording time, the frequency during playback will be the first
As shown in Figure 1, the time T is divided into m equal parts, and the time T is divided into m equal parts, and the time T is divided into m equal parts as shown in Figure 1. The signal level of the number printer signal is written into the storage means every time. That is, a reproduced signal existing for a time TK is subdivided into m 1m K subdivisions, and the subdivision signal levels are sequentially stored. Furthermore, mO
According to the sampling theorem, it is known that if the value is ≧2f in relation to the recording/reproduction frequency f, the original signal can be reproduced sufficiently clearly. Next, the m subdivision signals written in the storage means as described above are
As shown, when read every T/m-1/n time, T/n
m signals can be read out in time. That is, the time interval T (for a stored good signal) is compressed by T/n K. Since the sine wave existed for T f/n periods for the time width T of the reproduced signal O, that is, for each sampling signal, the time T of the good signal read out from the storage means is
/n ORK also has a sine wave for Tf/n periods. Therefore, the frequency of this read signal (e) becomes Kf as shown in Figure 1 (e). In other words, if the playback time is n times the chime time, then the time to read from the storage means is 1/n of the time to write to the storage means. Therefore, only the playback time can be extended or extended without changing the pitch of the sound during recording. When compressed, it becomes sharp 〇The fundamental frequency of the human voice Q is approximately Zoo -170- (
periodic 6-10m sea), approximately 250-5o for females
o-(periodic 2-4 m sea). In order for the basic scrap wave number to be counted as a pitch, it is necessary for waves of a certain period to repeat and continue for an appropriate number of times, and this IB number has a frequency of 100 to 50.
0 IIs OIE IB Te11:ias degree.

Therefore, when storing in the storage means, O? Convert printer time to reading time T/n K, T/n =
30 ms@e Set to 8 degrees) Enables expansion or compression of the human voice. Let us now consider the case of playing back an audio recording in a band of up to 10. If the reproduction is performed for a time n times the recording time, the maximum frequency f of the reproduction signal will be 10 x 10'' 1/nHs.

Here, sampling constant II m/T≧2f, a/T−
Considering the 2f4D case, f tf-10X103・1/n
If tT/* is set as T/n = 30X1 G' sea K and is 9, then m = 2 X (10XI 0-X1/n) X
Twx 2 X 10X10"X1/!l X 30
x 10' X n 600. That is, when the latter method is implemented as a device, at least 600 memory elements are required as the memory means.

Conventionally, it has been almost impossible to incorporate a large number of memory elements as described above into a tape recorder or the like at a low price of about 11 kiloliters. However, in recent years, with the development of IC technology, large-capacity analog memory elements have been developed, and packet pre-gate devices (hereinafter referred to as BB), which are one of the charge transfer devices, have been developed.
D) or charge-coupled devices (COD) have been put into practical use. %KBBD requires several hundred samples, s/m ratio 6 G d B, Xl[#JLO, PA5EsM, low manufacturing cost, and can be assembled in extremely small size.

Therefore, as will be described later, it is desirable to use the BBD as the storage means in the playback device of the present invention, and by doing so, the playback device can be fully incorporated into general audio equipment. .

Next, in a reproducing device that belongs to the latter method mentioned above, it is necessary to reproduce the recorded signal at a speed different from that during recording, and to make the pitch of the reproduced signal the same as the pitch of the recorded signal. It is necessary that the following relationship K fl m nfllk be established between the zero frequency f of the clock pulse when writing to the storage means and the frequency f of the clock pulse when reading from the storage means. n indicates the time ratio between recording time and playback time; t9 In the playback device, a clock pulse with an oscillation frequency f8 and a clock pulse O with an oscillation frequency f8 are generated separately. At least two O oscillators are required.

- Well, if the recorded signal O is to be reproduced while maintaining the relationship of f and 'x Hf, and the time ratio n can be set arbitrarily, it is assumed that the frequency f is constant and the number of revolutions f is varied. ``How to make the revolution number f constant, and the revolution number f,
There are two possible methods of changing method J. However, the time ratio nO setting and frequency fs (or 1.)
In practice, the 0 setting is often done using a variable resistor, etc. In that case, nine variable resistors are usually operated to set the f.
It is very difficult to maintain the 8wt nfl relationship and it is technically difficult to continuously change the set value. Therefore, the expansion and compression of the playback time cannot be set arbitrarily and continuously.6 To solve this problem, the frequency fx (or f,) is set to the playback speed of the recording medium such as tape. It is conceivable to automatically control the At this time, if the rotation number f1 is kept constant and f is automatically controlled according to the reproduction speed of the recording medium, the sampling time T becomes constant. Therefore, the above-mentioned fundamental frequency can be heard as a pitch if the necessary repetitions of the fundamental wave are not sampled, or if completely different shapes are sampled at the same sampling time TK. As a result, the reproduced sound becomes extremely unnatural and difficult to understand. The present invention has been made in view of the above points. That is, when converting the reproduction device according to the present invention into a recording medium, the clock pulse frequency f at the time of reading is kept constant, and the clock pulse frequency f at the time of writing and the reproduction speed of the recording medium are changed in relation to each other. By doing this, the necessary fundamental wave that can be heard as a pitch is reliably sampled, and the reproduced sound becomes extremely natural.

An embodiment of the present invention will be described in detail below with reference to the drawings. $211 is a configuration block diagram of an embodiment of a playback device according to the present invention. 1 is a magnetic tape for recording and reproduction as a storage medium, 2 is a capstan, and 3 is a pinch E2-1'
+, 4 are a tape supply reel and a 5Fi tape take-up reel, and the tape base is run by rotating the capstan 2 in accordance with the rotation of a motor 7, which will be described later. 6
is a reproducing magnetic head that is magnetically stored on the tape, detects a large signal, and converts it into an electric signal. 7 is a DC motor with an electronic governor, and the -80 electronic governor circuit controls the rotation speed. As a driving means for the reproducing device using the controlled zero-blow generator 11, it is quite possible to use another motor etc. for the DC motor with an electronic governor.◇9 is the motor B of the motor 7.
@It is a rounded variable resistor that changes the number. The motor 7 drives the capstan 2 (shaft 7m, pulleys 7b, 2e, and belt 7dK). 1G is an opaque disk having slits at equal intervals on its circumference, and the axis 7 of the motor
m.

Via pulley 7e and bell) 7f, pulley lOg,
The motor 7 is set to rotate at an increased rotational speed. 11 is a lamp, 12 is a round light-receiving element, for example, 7 ototo 2 digits, and the disc 10 rotates to form a slit. When the light from the lamp 11 comes on the line connecting the lamp 11 and the light receiving element 12, it receives the light from the lamp 11, converts it photoelectrically, and gives its electrical output to the input of the amplifier 13. Reference numeral 13 denotes an amplifier that amplifies the output of the light receiving element 12 and generates a pulse signal 13a'. 14 is an audio preamplifier; 015 and 16 amplify the reproduction signal from the reproduction head 6;
is the sampling of the reproduced signal and the memories 17 and 1, which will be described later.
80 electronic switches for circulation of contents, comprising: 11
1b, which are switched by the outputs of 7 lips, 7 g, and 2s, which will be described later. ◎ Front terminal 15
m, 16m are connected to the preamplifier 140 output terminal, 15b, 1
6b are connected to output terminals Kll of memories 17$P and 18, which will be described later, respectively. 17Th and 1$ are rounding memories that store each level of the subdivided large signal, and B
It is composed of O charge transfer devices such as llD.

19 is a round electronic switch for switching the output of the memories 17 and 18, and has terminals 11m and 19b on the input side, and 020 is a tape switch that is switched by the output of the 7rip 7w knob 28, which will be described later. An audio amplifier with an O equalization circuit to compensate for the frequency difference required during playback in a recorder, 21 is a variable resistor for volume adjustment, 22 is an audio main amplifier, and 23 is a loudspeaker. Ru. 24 is a pulse oscillator that oscillates at a constant frequency. 25 and 26 are step counters, which are reset by a 1kh electric circuit (not shown) at the start of running of the tape 1. The counter 25 receives the output signal of the amplifier 130, and the counter 26 receives the output signal of the oscillator 24, respectively. , the signal 25 m when it reaches a certain value.

26a, and continues to hold the output signal until a counter reset signal, which will be described later, arrives. 27' is an AND circuit which outputs both the output signals of 25m and 26m and outputs the 9 and 27m output signals 27ati counter 2;
The port 28 which is applied as a counter reset signal to each of 5 and 26 and serves as a trigger pulse for the 7-lip flop 28 is a 7-lip 70-tub, and when the tape 1 starts running, its output 28a is output by an electric circuit (not shown).
becomes the L level, and the AND circuit 270 output signal 27a
From K, the state level of the output is L→H→L→H→...
The output signal 28m of the 0#7 rip hoop 28, which changes sequentially as shown in FIG.
and controls switches 30 and 31, which will be described later.

Reference numeral 29 denotes a gate circuit, which opens the gate and inputs the output signal 13m of the width converter 13 when there is no output 25m of the counter 25, and inputs the input terminal 30m of switches 3o and 31, which will be described later.
We supply 31aK. (1829 can be omitted) 03° and 31 are clock pulse generators 32° and 330, which will be described later. An electronic switch 30a as an input terminal.

Has sob, sp and 31m, 31b, 7 lips 7a
It is controlled by the output signal 28sK of the knob 28. The terminals 30m and 31m are connected to the gate circuit 29, and the terminals 30b and 31b are controlled by the output signal 28sK of the oscillator 24. It is continued. Reference numerals 32 and 33 are clock pulse generation circuits that generate lock pulses necessary for the charge transfer device O transfer used in Memo 917 and '11 from one output pulse of switches 30 and 31, respectively; Since the BBD is used as a charge transfer device, two-phase clock pulse trains for controlling the accumulation mode and transfer mode are generated and supplied to the memories I717 and 18, respectively. As described later, the clock pulse generator 3
The clock pulses generated by 2.33 are sampled to subdivide the nine signals.

7 lip 70 tube 28 output 28 m and each switch 151
Regarding the relationship with the states of 16, 19, 30, and 31, the output 28a of the 7 lip-flop 28 is at L level -
The output of the G-DR 7 RATCH 15 is the input terminal 15 a K,
The output of switch 16 is 11bK, and the output of switch 19 is 19bl (.

The output of switch 30 is connected to 30&, the output of switch 31 is connected to 31b, and when 28a is at H level, the output of switch 15 is connected to 15b, the output of switch 16 is 16aKs, the output of switch 19 is 19JLK, and the output of switch 3o is connected to 15b. The output is 30b, and the output of switch 31 is 31aK.
are connected to each other. In the above configuration, now the tape l
Let the tape speed at the time of storage be ma, the tape speed at the time of playback be h, that is, the playback time be the recording time n. The tape speed during playback is determined by variable resistor 9. In particular, by adjusting the speed, the speed can be varied continuously.

In addition, in the following explanation, the 9-number alphabet attached to each terminal also means the signal input to or output from it, and conversely, the 9-number alphabet attached to each signal means that the signal is input. , or also means an output terminal. The highest frequency of the audio band O to be distributed and reproduced is f=1. When the time required to repeat the number of WAs of the base liquid necessary for the audio signal to be heard as audio is T, the oscillation frequency of the oscillator 24 is lf, (
=1m><1≧2), and the counts of counters 25 and 211 are respectively set to lfm% (that is, counters 25.
'26 starts counting from the reset state and reaches 1bT3
When nine pulses are applied to the input signal 25a.

26 m), the storage capacities (number of elements) of the memories 17 and 18 are lf and T, respectively, and the number of disk XIO slits is set as follows. That is, when the tape speed during reproduction is the same as the tape speed during recording, the frequency of the light from the lamp 11 received by one optical element is set to be 11-. Furthermore, when the tape 1 starts running, the counters 25 and 26 are set to 7 by an electric circuit (not shown), and the output 2'8 of the flip-flop 28 is
It is assumed that the iL level is reached and the contents of the memories 17 and 180 are cleared. As the tape 1 runs at the speed h, the disk 10 also rotates, and the light receiving element 12 receives a frequency of 1i4.
A pulse of 1/n(-f*) is generated, amplified by the amplifier 13, and applied to the counter 25 and the gate circuit 29 as a pulse signal 13m as shown in FIG. 3(0).

The counter 25 counts the Mukuro pulse signal 13m, and the third
As shown in Figure HK, K lfmT@ ・n/lfg = n
The signal 25m is output from the time T1 has elapsed and the count of lfmT@ is completed until the counter 25 is reset. Gate 29 is wrinkle signal 25
The pulse signal 13m is supplied to the input terminal 30& of the switch 300 and the input terminal 31aK of the switch 31 until m is output, that is, for a time nT1.The number of pulses during this period is 1fII/n X nT@ = 411 Ts
It is a preparation. The oscillator 24 generates a pulse with a frequency /f as shown in Fig. 3←), and the pulse is output to the input terminals 30b and 31b of the death counter 26, the switch 3o, and the input terminal 31bK.
Supplied. As shown in Figure 3), the counter 26 receives a signal from the time when one input pulse KIf-〒0 is added, that is, when it starts counting, until the count is reset after Ts・Vtf-=Ts time has elapsed. 2
11a is output. The moment when both the signals 25m and 26a are output is shown in FIG.

"The signal 27m from the urchin AND circuit 27 is output, and 7
The lip 70 tube 28 is inverted and the level of its output 211m changes from L to H. n-1 bakinsha output signal 25
1 and 26m occur at the same time. If Otori〉1, signal 2
After 5 breaths, KS4It No. 26a occurs first. If n<1, the signal 25m is generated first and the signal 26m is generated later.

First, the operation when vs>1 will be explained. The fMs diagram is n>
1 shows the operating liquid form of each part in FIG. 2. Since n>1, when the signal 25m is output, that is, when nT time has elapsed after the counter 25 is reset, the output signal 27& is generated in the AND circuit 27, and the output signal 28m of the 7-rig flop 28 is inverted. At the same time, counters 25 and 26 are reset.

Here, it is assumed that the signal 28m becomes L level when inverted. Assume that an audio signal whose fundamental frequency is f when reproduced at the same speed as the tape speed at the time of recording is recorded on tape l.

The reproduction signal (its fundamental frequency is f/n) detected by the reproduction head 6 is amplified by the preamplifier 14 and sent to the input terminal 15 of switches 15 and 160, respectively.

Supplied to 16m. Since the signal 2g& is at the L level, the input terminal of the memory 17 is supplied with the reproduction signal from the reproduction head 6, and the input terminal of the memo 918 is supplied with the output signal of the skeleton memory 18 in circulation. During this period, the reproduced signal is sampled by the switch 15, and the memory 17 also contains the signal 13, which has passed through the gate circuit 29 and switch 30, and the clock pulse (frequency 1fg/n
) is supplied, and the reproduction signal level is stored every n//fg time and sequentially transferred in the memory 17. In other words, the reproduction signal (sampling signal) sampled by the switch 15 is supplied from the clock pulse generator 32. The signal is subdivided by the clock pulses generated and stored in the memory 17. On the other hand, the memory 18 has a switch 31
The clock pulse (frequency ttg) generated by the oscillator 24C) output signal from the
) is supplied, and its contents are transferred in the memo 18 every 1/zrg time. The output signal from the memory 18 is supplied to the input terminal 16b of the switch 16, and again to the memory l.
It is sequentially transferred to
1, the volume is further amplified by the main amplifier 22, supplied to the loudspeaker 923, and converted into an audible signal. Since the counter 25 generates the output signal 25m when lfgT pulses arrive at the input side, the contents of the memory 17 are transferred lfgT times. Since the storage capacity of the memo+717 is rfgTt, when the contents of the memory 170 are all rewritten, the 7-rip 770 knob 28 is inverted and becomes H level. When the output signal 28a of the clip 70 becomes H level, the playback head 6 is connected to the input terminal of the memory 18.
The reproduction signal from the memory 170 is supplied to the input terminal of the memory 170, and the output signal of the circuit 17 is supplied to the input terminal of the memory 170 in a circulating manner. During this period, the reproduced signal is being sampled by the switch 16.

Furthermore, in the memory 18, a gate circuit 29 and a switch 31 are connected to a clock pulse generation circuit 3 by a signal 13m.
Clock pulse generated at 3 (frequency jfg/n
) is supplied, and the reproduction signal level is stored every hour of B/lf and sequentially transferred in the memory 18. That is, the nine reproduced signals (t sampling signal) sampled by the switch 16 are sent to the clock pulse generator 33.
and stored in memo 918. Oscillation 1B2 via switch 30
Based on the output signal of 4, the clock pulse generation circuit 32 generates 9 clock pulses (frequency It, ), and
1/lf The contents are sequentially transferred through the main amplifier 170 every hour, and are amplified by the equalizing amplifier 20 via the switch 19, the volume is adjusted by the variable resistor 21, and the content is transferred by the main amplifier 22. The signal is further amplified and supplied to the loudspeaker 23, where it is converted into an audible signal.

That is, the reproduction signal (third wJ()) of the preamplifier 14 with the fundamental frequency f/n is alternately supplied to the memories 17 and 18 every 18 hours, and the clock generated by the clock pulse generation circuits 32 and 33 is also supplied to the memories 17 and 18 alternately every 18 hours. 7fs by pulse
'rt is subdivided into equal parts and stored in the file 4917.18. The reproduction signal with the fundamental frequency f/n is 1, and the fundamental wave exists for a period of tT during nT1 time. That is, regardless of the ratio n of the storage time of tape 1 during recording and the playback time of tape 1 during playback, the memory 17-, 1
In g, the basic technique of the audio signal is fT, and the period is 1fg.
When it is subdivided into T% equal parts and stored, it becomes. When reading, the contents of the memory 17.18 are transferred every hour, so the fT of the fundamental wave is 1 hour T.

The period can be read out, and the frequency of the fundamental wave of the read nine signals is f. Since n>1 here, n
If T1>Tl, the writing time to the memory 17, 18 becomes longer than the reading time. Then, once the signal is read out, it is converted into an audible signal and at the same time it is memorized again.
J 17 , 18 is supplied to the input side of the memory 17.1
Therefore, during the time nT1 until the output signal 28m of the 7rip 70 tube 28 is inverted, the audio reproduction signal is repeatedly reproduced with its fundamental frequency set to f. As described above, by alternately putting the memories 17 and 18 into the writing and reading state in synchronization with the inversion of the flip-flop 28, the reproduced signal does not change the pitch at the time it was recorded, and its time period is expanded and the playback signal is sequentially expanded. It is being regenerated.

Next, the operation in the case of K n < 1 will be explained. In this case, the output signal 25a of the counter 25 should be generated before the output signal 26& of the counter 26 even if the counters 25 and 26 start counting at the same time.

Since the gate circuit 29 is not closed until the signal 25m is generated, the gate circuit 29 remains open for nT1 hours. The number of pulses supplied from the amplifier 13 to the switch 3 or 31 during this period is If-s T*.
So0JII Monkasha 1fm/7g.

At the time when the signal 26 is output, that is, the counter 25,
211 starts counting and T time elapses, and at the 9th point, the output 27m occurs in the AND circuit 27, and the counter 25
. Here, since n<1, mT1<Tl, and the write time to the memory 17.18 is the read time. At the time when fT and the period of the fundamental wave have been read, writing for the fT and period of the fundamental wave of the next signal has been completed in the other memory 18. Therefore, by reversing the 7-rip 7-o knob 28, the read/write state of the memory 17 and 18 is also reversed, and the playback signal of the tape is stored, and only that time is compressed and played back sequentially without changing the pitch at 9 o'clock. I'm going. The operations of other parts are exactly the same as in the case where n>1.

As mentioned above, in memo 1717.18, the fundamental wave of the audio signal is divided into equal parts and stored, where fT is the time ratio between recording and reproduction, nKm, and the period is jfg 'r, where n is It can be any continuously variable amount. Further, the period fT corresponds to the number of repetitions of the fundamental wave necessary for the fundamental frequency of the voice to be heard as a pitch.

In the above embodiment O, a disk 10 that rotates in conjunction with the motor 7 and has slits at equal intervals around the circumference is used as a rounding detection means for detecting the playback speed of a recording medium such as a tape. Lamp 11 and light-receiving element 1 sandwiched together
2, when the circle 1 [10 rotates in conjunction with the motor 7, that is, in conjunction with the running of the tape 10, the light from the lamp 11 that passes through the slit and reaches the light receiving element 12 is transmitted to the light receiving element 12. The charge is converted into a suitable detection signal for playback of the Zengji brass medium.

The wrinkle detection means is not limited to the above-mentioned embodiments, but it is also sufficient to record an audio signal and a 1-meter control signal on a recording medium such as a tape, and use these as detection signals. Achieve a goal.

Figure 4 is a diagram showing a detection means for detecting the speed of the recording medium by this method.

4 figures KX on 11? 2, the same parts Ka as shown in FIG. 2 are given the same reference numerals, and parts other than the detection means are omitted. 34 is a rounded control signal playback head that detects the 9 control signal recorded on the tape IK during recording, 35 is a rounded amplifier that amplifies the 9 control signal detected by the yi playback head 34, and the output haiku is rectangular. Wavy. 36 is a differentiating circuit for creating a differential waveform from the output of the rectangular liquid amplifier; 37 is a double-wave rectifier circuit, which generates a pulse signal 371 having twice the frequency of the detection signal from the output of the differentiating circuit; The tape 1 to be generated has a track for recording and reproducing control signals in addition to a track for recording and reproducing audio signals, and the wrinkle control signal is generated at the same frequency as when the tape 1 is reproduced at the same speed as when it was recorded. ◇The control signal 0 frequency @te is recorded so that it becomes lfm/2 without providing a special recording track for the control signal 0 most signal. By setting it to , it can also be used for a single track. In this embodiment, the detection signal is reproduced by the detection signal reproducing head 34, amplified by the amplifier 35, and then converted into a pulse signal 37 having twice the frequency by the differentiating circuit 36 and the double-wave rectifier circuit 37. According to the sampling theorem mentioned above, this must be true if the number of sampling signals 0jll#IL is 凇, the frequency of the original signal is f, and the relationship fb≧2faall is satisfied. Based on ◇ and 411
When recording and reproducing the detection signal using magnetic tape as a recording medium, the detection signal must be stably recorded and reproduced at a frequency that is at least twice the highest frequency of the audio signal.
Since reproduction is often difficult, the circuit configuration described above makes the number of sampling cycles of the detection signal twice the number of original cycles.

In the above configuration, since the intervals between the detection signals on the tape 1 are constant, the chirps also vary from tape 10 during playback.
If the running speed is 411 hours lower than the running speed at the time of recording, that is, if n>1, the number of cycles of the detection signal is 11Jn.
is the frequency If of the oscillator 24, which is very low, and when the running speed of the tape 10 during playback is suitable to the running speed during writing, that is, when n<1, the detection signal is 0J.
11 wave number If, / difficulty is oscillator 24014R number Jf.

Yop will also be higher. As the tape 10 runs, the detection signal is reproduced by the playback head 34, the detection signal is amplified by the amplifier 35, and converted into a pulse signal 37aK having twice the frequency by the differentiation circuit 36 and the double-wave rectification circuit 37, and then sent to the counter 25 and It is input to the gate circuit 29. The following operations are exactly the same as those in the previous embodiment.

In addition to the above-mentioned embodiments, for example, there is another example as a means for detecting a good signal proportional to the playback speed IILK of a recording medium such as a tape! If it is a round structure that detects an output proportional to the running speed of the recording medium, such as an assembly of a magnet and a reed switch, the present invention is suitable! It can be fully used as a detection means for 5-generation equipment. In the previous embodiment, we assume that the tape speed is only continuous, but in the case of a tape recorder whose speed can be changed to two or three speeds, the oscillation frequency of the oscillator 24 and the counter 25 are adjusted accordingly. This can be achieved by increasing the number of kakunts of .2- or by changing the speed increasing ratio between the motor 7 and the disk 10 when increasing the rotational speed of the motor 7.

In the above two embodiments, the case where the playback device according to the present invention is incorporated into a tape recorder has been described, but the present invention is not limited to this, and can be implemented in any recording/playback device such as a record player. It goes without saying that it is possible.

As is clear from the above description, the present invention is a playback device in which the time width of the signal that is finally expanded or compressed is kept constant, and the sampling time is varied depending on the signal related to the playback speed. , the repetition of the fundamental wave necessary for the fundamental frequency of the voice to be heard as a pitch is reliably obtained, and the reproduced sound becomes extremely clear.

Since nine expansion ratios or compression ratios are given continuously, the reproduced sound not only has the above-mentioned clarity, but also approaches natural sound, and can be applied to music, etc. in addition to the human voice.

Furthermore, if the above-mentioned BBD or the like is used as a memory, it can be easily incorporated into general audio equipment. It is connected after the switch 19. Before the frequency f/n of the reproduced signal is converted to the recording frequency, this equalization circuit is installed, for example, after the preamplifier 140. It is difficult to obtain sufficient thickening properties for continuous nO values. However, as in the above embodiment, if the frequency f/n of the reproduced signal is converted to the recording frequency fK so that it has equalization characteristics after the round, sufficient equalization characteristics can be obtained regardless of the value of n. It is something that can be easily obtained.

As described above, the present invention has many advantages and is highly effective.

[Brief explanation of the drawing]

Figure 1 is a diagram showing the principle of the present invention! ! FIG. 2 is a diagram showing one embodiment of the present invention, FIG. II3 is a diagram showing waveforms of various parts when n>1, and FIG. 4 is a diagram showing a further embodiment of the detection means. l...Magnetic tape for recording and reproduction, 6...
Reproducing magnetic head, lO...Disc with slit, 1
1... Lamp, 12... Light receiving element, i5, 16, 19, 30, 31... Switch,
17.18...Memory, 24...
・Oscillator, 32.33...Rep pulse generation circuit, 34...Detection signal reproducing head 136.
...Differential circuit, 37...Double wave rectifier #l direction path.・Patent applicant Canon Co., Ltd.

Claims (1)

[Claims] a reproduction means capable of reproducing a signal recorded on a recording medium and changing the reproduction speed; a signal generating means generating a signal associated with the reproduction speed of the scissors reproduction means;
a sampling means for sampling the signal reproduced by the amplifier means; a storage means for dividing and storing the signal sampled by the amplifier means according to the signal generated by the signal generating means; and output means for outputting according to a reference signal having a predetermined constant frequency, the playback device expands or compresses only the playback time without changing the pitch of the sound during recording. A first counter that counts the generated signal, a second counter that counts the reference signal #J, and an AND circuit that receives the output signals of the first and second counters as input. The sampling time of the number printer means is controlled according to the output of the AND circuit, and the number of count stages of the counters of the clusters 1 and W, 2 and the number of storage elements of the storage means are made to match. Regeneration bushes 0