Is it posible to realize voice communication using unlicensed frequency spectru

I want to do my main project in this domain.
The motive behind the poll is to inquire whether voice communication is viable or not without using the licensed frequency spectrum.

Of course it's possible. Skype VoIP over 802.11a/b/g is all the proof you need.
Any cordless phone is also doing 100% of what you describe.
A lot of cheap on-off-keying radios can manage 9,600 BPS which is enough for GSM CODEC audio.
I doubt that the Arduino has enough processing power, memory or I/O bandwidth to actually do audio compression of this type though.

You could certainly do it over wi-fi. I'm not sure it really plays to arduino's strengths...

The low power toy "walkie-talkies" do it all the time.

Thanks for your suggestions.
As we see it is ubiquitous sending the data files via bluetooth and zigbee, i just want to explore whether full duplex voice communication works on zigbee as its range is much higher than bluetooth.
I also want to know about the new technologies having larger ranges than existing and supporting the above said demands.

i just want to explore whether full duplex voice communication works on zigbee

No, not enough bandwidth.

You are being very vague.

It depends on what you want to do. Do you want to build your own transmitter or do you want to use ready built modules?
A lot depends on where about in the world you are as the regulations differ from country to country. Most places don't allow you to make transmitters, even on the license free bands. These transmitters must be type approved. That is the design must be tested in a recognised test house.

Take an example of Bluetooth powered headsets. They come with dual functions (microphones and earphones) enabling us to get access both to hear and talk.
So, if we try to extend this concept to zigbee/xbee, then its like we have achieved wireless voice transmission between two devices placed much far than bluetooth.
What do you say?