The best thing I know is to use TRAX software from Flux / IRCAM, for which an astonishing video is posted here. The software enables independant Pitch / Formant / Gender / Age control as well as mapping to subtle LFOs, and ability to add further Formant/Spectral envelopes. The software is designed for both speech and singing voice signals.
The easiest way however is actually just to use Pitch Shifting, which is the most available type of audio processing. Although you might know what the original Pitch shift setting was, doesn't mean it is easily reversible or recognisable from another's perspective.
A combination of EQ colouring between 400hz-4khz (Peak/Notch filters) and then pitch shifting could present a more unique result.
Frequency Shifting and Ring Modulation are very limited and can drastically affect the speech intelligibility.
You could try Vocoding software, although the idea of Vocoding is to really use the formant information already present in the voice signal and substitute the 'vocal cords' for another type of vibration (e.g. another source, saw/square wave). This really transforms the sound a lot more than just the 'identity' that you might be referring to, since the identity of one's voice is a product of the entire physical/vocal system, basically all dimensions of one's head, neck, throat, mouth, larynx and vocal cords, etc, which affect the sound a lot more subtly.