The motor is an old one from a cassete player of some sort. I am powering the motor with a 9 volt battery, but it's RPM is too high for the desired purpose. I have used gears to reduce RPM but it is still too high. I don't think I can increase the gear ratio anymore. Can you suggest a way to decrease the RPM further and increase the torque (if possible). Please suggest a circuit to accomplish it.
I shall make the assumption the machine you have is a brushedDC machine.
Torque ~= current (Kt being the constant of proportionality)
Speed ~= voltage. (Ke being the constant of proportionality)
In actual fact this is loosely accurate for all machine types but rather than being directly proportional it becomes a bit more complex & usually involves frequency. ( see How do DC motors work with respect to current, and what consequence is the current through them?)
To reduce the speed the simplest method is to reduce the voltage be it by replacing the 9V battery by a... 8V battery or via PWM.
To increase the torque ... that a bit harder. All you can do, electrically (as you can do it mechanically via gears), is to reduce the source impedance to maximise the current that can be delivered for a given stator impedance.
Rewind it with more turns of thinner wire. For a given voltage, speed * turns will be a constant, so quadrupling the turns would give 1/4 the speed. To fit those turns in the same space, you would need to halve the wire diameter.
Gearing is easier : here will be ways of increasing the gearing : the simplest way to get a large speed reduction is using worm gears.
But you haven't told us anything about how much speed reduction you need.
To decrease the RPM at the same torque, you want to power it from a "stiff" voltage source that is lower. PWM is one way to achieve that with minimal losses.
To increase torque beyond what you get with 0\$\Omega\$ source impedance, you can go negative on the source impedance or use feedback to actively adjust the voltage to maintain speed (more complex and you need to get the control loop right).
I think you could do simple negative source resistance (also called IR compensation) with an op-amp driving the reference terminal of a linear regulator, which might be appropriate for a small motor.