I have had success with two methods of pitch shifting a V-Tech toy.
The first method required relocating the coil (chrome cube) from the PCB to a hole I made in the body of the toy. With the coil exposed I could pass magnets of differing strength near the coil to speed up the pitch. The coil creates a magnetic field when current passes through, so disturbing that field with another magetic field increases the pitch. To be able to have a lower pitch I installed a second and bigger coil. I use a switch to choose which coil to use, and this gives me slow, normal, and fast.
The second method involves removing some components, and adding on an oscillator. If there are 2 traces that connect 2 pins of a V-Tech chip to ground through a coil and one cap per trace that is an oscillator. The oscillator is supplying a steady clock signal to the chip. Replacing that oscillator with an adjustable one will allow for pitch shifting. Try different kinds of oscillators and different values for resistors, pots, and caps. I usually start with an astable 555*. If that does not do what I want I try a 4011 based oscillator. The 4011 is just 2 inverters, 1 res, 1 pot, and 1 cap. Any chip that you can make inverters with (including a hex inverter chip) can be substituted. So far this method has given a range of normal to super slow.
*I like the 555 astable because if it does something interesting it can usually be expanded by adding a second astable 555, into pin 5 of the first 555, as a LFO.
Anyways so you make your oscillator. Now remove the coil and the 2 caps. Then probe each of the two traces to see which one you will attach your new adjustable oscillator to.
This is the schematic of the 4011 I am testing right now. I am still trying different values to try to get higher pitch than the normal. I put an LED on the output to have a visual reference of the rate. With this osc the rate is so high that the LED glows steady, but when I removed it the toy crashed. Just leaving a 1K res fixed the problem. I tie in the power and ground into the toy's circuit where the toy's power switch will also turn on/off the osc. Otherwise the osc will drain the batteries. The open lug of R2 can either stay open or be tied to the middle lug.
The values I have had the best results with (for Alphabet Desk) are:
R1 100 ohms
R2 25K pot
C1 .01 uF
Update: It seems as thought the chip on the Alphabet Desk has OSC OUT and OSC IN pins. A schematic of a stable pitch mod can be found on the bottom of the Alphabet Desk page (the pitch mod is pins 1-4 of the 4069).
The V-Tech Alphabet Desk
Here are photos of the circuitboard with the coil and caps removed. There is a wire to each trace for testing.
Tried to make the traces more obvious. The metal case of the coil is grounded to shield the coil and rest of the circuit from any interference.
Here are photos of the first method using magnets for pitch shift. I have a the magnets mounted on a plastic rod to make them easier to hold.
My bent toys and homemade circuits.