Saturday, January 21, 2012

A digital voice faker

This device is a generator based on the voice audio chip from Holtek HT8950, when talking into the microphone, the voice from the loudspeaker can be altered as desired alone or together using the three available effects: change of tone, vibrato and robotic voice.

 Figure 1: Diagram of Digital Voice faker.

The assembly is based on the HT8950, a component that operates a shift upward frequency (one octave highest) and down (up to 2 / 3 of octave lower) in six steps: three up and three down.
The frequency translation is the heart of the generation of voice and is already making an unrecognizable person to listen, it also allows to simulate well enough votes "original" of the kind of cartoons or movies horror or science fiction.
In addition, the chip Holtek has two other effects: the voice like metal robot voice vibrato, octave added to the displacement, these changes make the voice of a person truly unrecognizable. The operation mode can be set using the four buttons connected to ground or managing the corresponding lines (and TGD TGU and VIB and ROB) with logic levels 1 and 0.
Regarding the "shift", each press of the button determines the corresponding one step forward: towards acute TGU (pin 3, push SW3) and to the grave with TGD (pin 4, SW4 button). ROB (pin 5, SW5 button) inserts the robotic voice effect and VIB (pin 2, push SW2) vibrato, like the other two, these lines activate the corresponding function when they are set to logical zero, however, while ROB activated with a pulse at logic zero and turns off by going back with UP and DOWN control VIB is latched. In fact, the vibrato effect is activated by giving a pulse to pin 2 at logic zero and turns off with a second pulse, if you prefer, each switching 1 / 0 logic state opposite the vibrato. No reset command transparent mode (normal voice) was provided by the manufacturer, however it is possible to return by pressing SW3 (or UP) and SW4 (or DOWN) or by depriving the integrated circuit supply and feeding it again.
At power up, the integrated circuit automatically activates the robot effect, to return to the original, repeat three steps up (with the UP button, pin 3) or down (with the push DOWN pin 4). This is because the scale of effects we can achieve with the buttons, the robot voice in mind is to say over the alteration most acute for when you press the pusher robot voice (pushbutton connected to pin 5), HT8950 control logic automatically turns off the frequency shift: just a step above. Here's why, to return to the frequency shift or normal voice, he must take four steps up (four pressures UP) or down (four pressures DOWN).
Management using the buttons is cyclical: we start from the normal position, it releases more than three times the UP or DOWN and mode of operation continues to change. In fact, after three steps up (UP) providing the integrated circuit for the robot voice (even without pressing the Robot) and therefore, if you press UP again, the IC turns voice dragging to the grave, to return to normal voice (no effects added). Same for DOWN, except that after the pitch corresponding to the maximum slip down (2 / 3 octave lower), we still find the robot voice, so that the more acute, until returning to normal voice.
The vibrato effect is an additional effect that can be added to the robot voice, so to get it he must first activate the voice robot (by pressing the UP or DOWN buttons, or pressing immediately pusher robot) and press the button VIB (connected to pin 2). This effect is a frequency modulation of the voice signal.
When you press the button VIB voice glides in frequency by one step up and down alternately, at a frequency of 8 Hz
To build this digital voice faker you will not encounter any particular difficulty, even if you're a beginner. Then the camera will take place in a small plastic housing. All components and circuit board are available from some of our advertisers, so no supply problem.

 Figure 2: Photograph of a prototype of the trickster's Platinum digital voice.

Iist EV171
R1 ...... 330O
R2 ...... 330O
R3 ...... 470O
R4 ...... 100 kO
R5 ...... 47 kilohm
R6 ...... 4.7 kO
R7 ...... 4.7 kO
R8 ...... 10 kW
R9 ...... 10 kW
R10 ..... 1.2 kO
R11 ..... 100O
RV1 ..... 47 kilohm trimmer
RV2 ..... 220O pot linear
C1 ...... 100 nF ceramic
C2 ...... 100 nF ceramic
C3 ...... 100 nF ceramic
C4 ...... 100 nF ceramic
C5 ...... 22 uF 16 V electrolytic
...... C6 4.7 uF 16 V electrolytic
C7 ...... 10 uF 16 V electrolytic
C8 ...... 100 uF 16 V electrolytic
C9 ...... 100 uF 16 V electrolytic
C10 ..... 100 uF 16 V electrolytic
ZD1 ..... 3 V 0.5 W zener
LD1 ..... LED 3 mm red
LD2 ..... LED 3 mm red
IC1 ..... HT8950A
IC2 ..... LM386N
MIC1 .... microphone capsule
SW1 ..... slide switch
SW2 ..... micropoussoir
[...]
SW5 ..... micropoussoir

Miscellaneous:
HP 8 ohm 1 W
9 V battery holder

Note: All resistors are quarter of W unless otherwise specified.

Potential Uses
As such, the trickster of votes is done to alter the voice on the phone or during a recording with a tape recorder with a microphone, in fact, if we're talking close to the electret, the subwoofer reproduces the voice changed.
On the phone, the speaker circuit is affixed against the telephone handset, for recording, the speaker should be closer to the microphone of the magneto. Of course it works even better when the capsule electret MIC1 is maintained at a certain distance from the speaker: it avoids the unpleasant feedback and, as the phone recording, we avoid short-circuiting the selected effect (which would be stupid!).
Apart from direct development, the trickster of voice can be used to produce the soundtrack to your video productions: in this case, disconnect the speaker and connect the output of the input audio or video recorder mixer, keeping the volume low to avoid saturating the line and get distorted recordings.

No comments:

Post a Comment

Most visited posts this week