Filtering and enhancement of voice calls in a telecommunications network

1. A method comprising:

receiving, by one or more devices, an indication, from a caller associated with a call, that speech of the caller is to be modified to deemphasize an accent of the caller;

modifying, by the one or more devices and based on the received indication, the speech of the caller to deemphasize the accent of the caller;

transmitting, by the one or more devices, the modified speech to a callee associated with the call;

receiving a second indication from the caller that speech of the callee is to be modified to deemphasize an accent of the callee;

modifying, in response to the second indication, the speech of the callee to deemphasize the accent of the callee; and

outputting the modified speech of the callee to the caller.

2. The method of claim 1 , further comprising:

receiving, from the caller, training data corresponding to speech of the caller speaking one or more model sentences or phrases; and

deriving one or more parameters from the training data,

wherein modifying the speech includes filtering the speech using the one or more parameters.

3. The method of claim 1 , wherein the received indication includes preference data defining one or more rules relating to when the speech of the caller is to be modified.

4. The method of claim 3 , wherein the one or more rules include a list of telephone numbers or names identifying callees for which the speech of the caller is to be modified.

5. The method of claim 1 , wherein the modification of the speech of the caller is provided as a service in a cellular network.

6. The method of claim 1 , wherein the modification of the speech of the caller further includes:

modifying audio associated with the speech of the caller to remove background noise.

7. A device comprising:

a memory device;

at least one processor to execute instructions in the memory to:

receive an indication, from a caller associated with a call, that speech, of the caller is to be modified to deemphasize an accent of the caller;

modify, based on the received indication, the speech of the caller, to deemphasize the accent of the caller;

transmit the modified speech to a callee associated with the call;

receive a second indication from the caller that speech of the callee is to be modified to deemphasize an accent of the callee;

modify, in response to the second indication, the speech of the callee to deemphasize the accent of the callee; and

output the modified speech of the callee to the caller.

8. The device of claim 7 , wherein the at least one processor is to further execute instructions in the memory to:

receive training data corresponding to speech of the caller speaking one or more model sentences or phrases; and

derive one or more parameters from the training data,

wherein modifying the speech includes filtering the speech using the one or more parameters.

9. The device of claim 7 , wherein the received indication includes preference data defining one or more rules relating to when the speech of the caller is to be modified.

10. The device of claim 9 , wherein the one or more rules include information identifying callees for which the speech of the caller is to be modified.

11. The device of claim 9 , wherein the modification of the speech of the caller is provided as a service in a cellular network.

12. The device of claim 7 , wherein the modification of the speech of the caller further includes:

modifying audio associated with the speech of the caller to remove background noise.

13. A method comprising:

receiving, by one or more devices, audio corresponding to speech of two or more participants to a telephone call;

modifying, by the one or more devices, the received audio to selectively enhance the speech of one or more of the participants to the telephone call, the modification of the received audio being controlled by one or more participants to the telephone call; and

transmitting, by the one or more devices, the modified speech to one or more destination participants to the telephone call;

identifying when a selected participant of the at least two or more participants is speaking; and

modifying the received audio, based on the identification of the selected participant, to enhance speech of the selected participant without enhancing speech of the non-selected participant.

14. The method of claim 13 , wherein the modification to the received audio to enhance the speech of the one or more participants includes deemphasizing an accent of the one or more participants.

15. The method of claim 13 , wherein the modification to the received audio to enhance the speech of the one or more participants includes modifying a tone of speech of the one or more participants.

16. The method of claim 13 , wherein the modification of the received audio is initiated in response to a command received during the telephone call and from the one or more participants to the telephone call.

17. The method of claim 13 , further comprising:

receiving training data corresponding to speech of one or more of the participants speaking one or more model sentences or phrases; and

deriving one or more parameters from the training data,

wherein modifying the received audio includes modifying the received audio based on the one or more parameters.