Granted Patent B2

US 9,105,053 · App. 13/634,811 · Granted Aug 11, 2015

Abstract

Claims (67)

1. A method comprising:

causing, at least in part, acquisition of voice data at an age estimator, wherein the age estimator comprises an input module and a computation module;

calculating a first set of probability values using the computation module, wherein each of the probability values represents a probability that the voice data is in a respective one of a plurality of predefined age ranges, and the predefined age ranges are segments of a lifespan;

deriving a second set of probability values using the computation module, by applying a correlation matrix to the first set of probability values, wherein the correlation matrix associates the first set of probability values with probabilities of the voice data matching individual ages over the lifespan;

for each of the predefined age ranges, calculating a sum of the probabilities in the second set of probability values corresponding to the individual ages within the respective predefined age ranges using the computation module; and

determining the predefined age range to associate with the voice data using the computation module based, at least in part, on the calculated sums of the probabilities.

2. A method of claim 1 , further comprising:

receiving an input specifying a target age range using the input module;

calculating another sum of the probabilities in the second set of probability values corresponding to the individual ages within the target age range using the computation module; and

determining whether the voice data is associated with an age within the target age range based, at least in part, on the another sum using the computation module.

3. A method of claim 1 , further comprising:

collecting a plurality of voice data training samples;

identifying each of the plurality of voice data training samples according to the predefined age ranges and the individual ages in the lifespan;

calculating a reference set of probability values for correlating the voice data with the predefined age ranges using the computation module; and

calculating the correlation matrix using the computation module, based on the voice data training samples.

4. A method of claim 3 , further comprising:

optimizing the correlation matrix via discriminative training using the computation module.

5. A method of claim 1 , wherein the calculating the first set of probability values, the second set of probability values, or both are performed using a statistical model including a Gaussian mixture model, an artificial neural network, a support vector model, a decision tree or any combination thereof.

6. A method of claim 1 , wherein each element of the correlation matrix represents a probability that one of the predefined age ranges falls into one of the individual ages.

7. A method of claim 1 , further comprising:

causing, at least in part, acquisition of biometric information, contextual information, or a combination thereof using the input module,

wherein the determining of the predefined age range to associate with the voice data is further based, at least in part, on the biometric information, the contextual information, or a combination thereof.

8. A method of claim 1 , further comprising:

selectively providing advertisements, recommendations, reports, notices, messages, media or any combination thereof, based on the determined predefined age range.

9. An apparatus comprising:

at least one processor; and

at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,

cause, at least in part, acquisition of voice data at an age estimator, wherein the age estimator comprises an input module and a computation module;

calculate a first set of probability values using the computation module, wherein each of the probability values represents a probability that the voice data is in a respective one of a plurality of predefined age ranges, and the predefined age ranges are segments of a lifespan;

derive a second set of probability values using the computation module, by applying a correlation matrix to the first set of probability values, wherein the correlation matrix associates the first set of probability values with probabilities of the voice data matching individual ages over the lifespan;

for each of the predefined age ranges, calculate a sum of the probabilities in the second set of probability values corresponding to the individual ages within the respective predefined age ranges using the computation module; and

determine the predefined age range to associate with the voice data using the computation module based, at least in part, on the calculated sums of the probabilities.

10. An apparatus of claim 9 , wherein the apparatus is further caused, at least in part, to:

receive an input specifying a target age range using the input module;

calculate another sum of the probabilities in the second set of probability values corresponding to the individual ages within the target age range using the computation module; and

determine whether the voice data is associated with an age within the target age range based, at least in part, on the another sum using the computation module.

11. An apparatus of claim 9 , wherein the apparatus is further caused, at least in part, to:

collect a plurality of voice data training samples;

identify each of the plurality of voice data training samples according to the predefined age ranges and the individual ages in the lifespan;

calculate a reference set of probability values for correlating the voice data with the predefined age ranges using the computation module; and

calculate the correlation matrix using the computation module, based on the voice data training samples.

12. An apparatus of claim 11 , wherein the apparatus is further caused, at least in part, to:

optimize the correlation matrix via discriminative training using the computation module.

13. An apparatus of claim 9 , wherein the calculating the first set of probability values, the second set of probability values, or both are performed using a statistical model including a Gaussian mixture model, an artificial neural network, a support vector model, a decision tree or any combination thereof.

14. An apparatus of claim 9 , wherein each element of the correlation matrix represents a probability that one of the predefined age ranges falls into one of the individual ages.

15. An apparatus of claim 9 , wherein the apparatus is further caused, at least in part, to:

cause, at least in part, acquisition of biometric information, contextual information, or a combination thereof using the input module,

wherein the determining of the predefined age range to associate with the voice data is further based, at least in part, on the biometric information, the contextual information, or a combination thereof.

16. An apparatus of claim 9 , wherein the apparatus is further caused, at least in part, to:

selectively provide advertisements, recommendations, reports, notices, messages, media or any combination thereof, based on the determined predefined age range.

17. An apparatus of claim 9 , wherein the apparatus is a mobile phone further comprising:

user interface circuitry and user interface software configured to facilitate user control of at least some functions of the mobile phone through use of a display and configured to respond to user input; and a display and display circuitry configured to display at least a portion of a user interface of the mobile phone, the display and display circuitry configured to facilitate user control of at least some functions of the mobile phone.

18. A non-transitory computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to at least perform the following steps:

causing, at least in part, acquisition of voice data at an age estimator, wherein the age estimator comprises an input module and a computation module;

calculating a first set of probability values using the computation module, wherein each of the probability values represents a probability that the voice data is in a respective one of a plurality of predefined age ranges, and the predefined age ranges are segments of a lifespan;

deriving a second set of probability values using the computation module, by applying a correlation matrix to the first set of probability values, wherein the correlation matrix associates the first set of probability values with probabilities of the voice data matching individual ages over the lifespan;

for each of the predefined age ranges, calculating a sum of the probabilities in the second set of probability values corresponding to the individual ages within the respective predefined age ranges using the computation module; and

determining the predefined age range to associate with the voice data using the computation module based, at least in part, on the calculated sums of the probabilities.

19. A non-transitory computer-readable storage medium of claim 18 , wherein the apparatus is caused, at least in part, to further perform:

receiving an input specifying a target age range using the input module;

calculating another sum of the probabilities in the second set of probability values corresponding to the individual ages within the target age range using the computation module; and

determining whether the voice data is associated with an age within the target age range based, at least in part, on the another sum using the computation module.

20. A non-transitory computer-readable storage medium of claim 18 , wherein the apparatus is caused, at least in part, to further perform:

collecting a plurality of voice data training samples;

identifying each of the plurality of voice data training samples according to the predefined age ranges and the individual ages in the lifespan;

calculating a reference set of probability values for correlating the voice data with the predefined age ranges using the computation module; and

calculating the correlation matrix using the computation module, based on the voice data training samples.

Assignments (10)