Computer-implemented system and method for assigning call agents to callers

1. A computer-implemented system for assigning call agents to callers, comprising:

two or more devices, each device comprising one of a conventional telephone handset and portable handset;

a database to store a list of traits and for at least one trait, two or more populations, wherein each population for one such trait is associated with a distribution of measurements from voice samples for the trait and at least a portion of the measurements associated with each of the populations of the trait are overlapping; and

a speech recognition server comprising a central processing unit, memory, input port to receive further voice samples from callers via one or more of the devices, and output port, wherein the central processing unit is configured to execute modules comprising:

an analysis module to analyze at least one of the further voice samples from one of the callers;

a trait module to identify traits of the caller from the analyzed voice sample, to determine that at least one of the identified traits comprises characteristics of each of the populations for that trait based on a value for the identified trait that falls within the overlapping measurements from the voice samples, and to define the at least one identified trait as a percentage of each population comprised in that trait;

a comparison module to compare the caller traits to traits associated with each active agent within the call center;

an assignment module to assign to the caller the agent with the traits most similar to the caller traits; and

a call distributor to route the call to the assigned agent.

2. A system according to claim 1 , further comprising:

an agent selection module to identify the agent assigned to the caller based on one of an absolute number of shared traits with the caller, a threshold number of shared traits, and a highest number of shared traits.

3. A system according to claim 1 , wherein at least one of the identified traits for the caller is determined as a binary value for a single population associated with that trait.

4. A system according to claim 1 , further comprising:

a speech recognizer to perform speech recognition on the voice sample and to identify one or more predetermined terms and phrases.

5. A system according to claim 1 , wherein the trait module determines at least one of the identified traits for the caller by identifying whether the caller fits more closely with one population than at least one other population defined for that trait and assigning the more closely fit population as the trait for the caller.

6. A system according to claim 1 , further comprising:

a coaching module to coach the assigned agent during the call with the caller, comprising:

a further trait module to identify further traits of the caller during the call;

a persuasion determination module to determine a stage of persuasion of the caller based on the further identified traits; and

a guidance module to provide guidance to the agent based on the caller's determined stage of persuasion.

7. A system according to claim 1 , further comprising:

an identity determination module to confirm an identity of one or more of the active agents prior to comparison of the agent traits with the caller traits, comprising:

an utterance module to receive from one such agent speech utterances;

a voice comparison module to compare the speech utterances with a voice print associated with the agent;

a similarity module to determine a degree of similarity between the voice print and the speech utterances;

a call connection module to connect the agent with the caller when the similarity is high.

8. A system according to claim 7 , further comprising:

a voice print module to generate the voice print by receiving a voice recording from the agent and generating from the voice recording a statistical model as the voice print.

9. A system according to claim 1 , wherein the traits each comprise one of comprehensibility, origin, accent, intelligence level, education, age, social characteristics, emotion, perceived appeal, personality, speech rate, gender, and competence.

10. A computer-implemented method for assigning call agents to callers, comprising:

storing a list of traits and for at least one trait, two or more populations, wherein each population for one such trait is associated with a distribution of measurements from voice samples for the trait and at least a portion of the measurements associated with each of the populations of the trait are overlapping;

receiving into a call center a call from a caller via a device comprising one of a conventional telephone handset and portable handset;

obtaining a further voice sample from the caller during the call via a speech recognition server comprising a central processing unit, memory, input port to receive the further voice sample from the caller via one or more of the devices, and output port;

analyzing via the central processing unit the further voice sample of the caller;

identifying traits of the caller from the analyzed voice sample via the central processing unit;

determining via the central processing unit that at least one of the identified traits comprises characteristics of each of the populations for that trait based on a value for the identified trait that falls within the overlapping measurements from the voice samples;

defining via the central processing unit the at least one identified trait as a percentage of each population comprised in that trait;

comparing the caller traits to traits associated with each active agent within the call center via the central processing unit;

assigning to the caller the agent with the traits most similar to the caller traits via the central processing unit; and

routing the call to the assigned agent.

11. A method according to claim 10 , further comprising:

identifying the agent assigned to the caller based on one of an absolute number of shared traits with the caller, a threshold number of shared traits, and a highest number of shared traits.

12. A method according to claim 10 , further comprising:

determining at least one of the identified traits for the caller as a binary value for a single population associated with that trait.

13. A method according to claim 10 , further comprising:

performing speech recognition on the voice sample; and

identifying one or more predetermined terms and phrases.

14. A method according to claim 10 , further comprising:

determining at least one of the identified traits for the caller, comprising:

identifying whether the caller fits more closely with one population than at least one other population defined for that trait; and

assigning the closely fit population as the trait for the caller.

15. A method according to claim 10 , further comprising:

coaching the assigned agent during the call with the caller, comprising:

identifying further traits of the caller during the call;

determining a stage of persuasion of the caller based on the further identified traits; and

providing guidance to the agent based on the caller's determined stage of persuasion.

16. A method according to claim 10 , further comprising:

confirming an identity of one or more of the active agents prior to comparison of the agent traits with the caller traits, comprising:

receiving speech utterances from one such agent;

comparing the speech utterances with a voice print associated with the agent;

determining a degree of similarity between the voice print and the speech utterances;

when the similarity is high, connecting the agent with the call center.

17. A method according to claim 16 , further comprising:

generating the voice print, comprising:

receiving a voice recording from the agent; and

generating from the voice recording a statistical model as the voice print.

18. A method according to claim 10 , wherein the traits each comprise one of comprehensibility, origin, accent, intelligence level, education, age, social characteristics, emotion, perceived appeal, personality, speech rate, gender, and competence.