System and method for speaker authentication and identification
A system and method for enrolling a speaker in a speaker authentication and identification system (AIS), the method comprising: generating a user account, the user account comprising: a user identifier based on one or more metadata elements associated with an audio input received from an end device; generating a first i-vector from an audio frame of the audio input, a trained T-matrix, and a Universal Background Model (UBM), wherein the first i-vector generation comprises an optimized computation; and associating the user account with the first i-vector.
1. A method for enrolling a speaker in a speaker authentication and identification system (AIS), the method comprising:
generating a user account, the user account comprising: a user identifier based on one or more metadata elements associated with an audio input received from an end device;
generating a first i-vector from an audio frame of the audio input, a trained T-matrix, and a Universal Background Model (UBM), wherein the first i-vector generation comprises an optimized computation; and
associating the user account with the first i-vector, and wherein the optimized computation comprises:
generating a feature matrix based on a feature vector;
generating a Gaussian Mixture Model (GMM) mean matrix based on a plurality of mean vectors associated with a plurality of GMM components; and
generating a delta matrix based on the feature matrix and the GMM mean matrix.
2. The method of claim 1 , wherein the optimized computation comprises:
generating a first multi-dimensional array comprising a plurality of duplicated matrices, each matrix comprised of a plurality of GMM mean vectors;
generating a multi-dimensional feature matrix, comprising a plurality of feature matrices, each feature matrix corresponding to a feature vector of a single audio frame; and
generating a multi-dimensional delta array based on the first multi-dimensional array and the multi-dimensional feature matrix.
3. The method of claim 1 , wherein the optimized computation comprises:
detecting diagonal matrices; and
performing only those computations that involve diagonal elements.
4. The method of claim 1 , wherein the optimized computation comprises:
detecting computations in an intermediate result that generate an off-diagonal element of a matrix which is diagonalized; and
eliminating the computation of the intermediate result.
5. The method of claim 1 , wherein the optimized computation comprises:
detecting a recurring computation;
precomputing the recurring computation; and
storing the precomputed result in a cache of a processor.
6. The method of claim 1 , wherein the optimized computation comprises:
detecting a computation between a first matrix and a second matrix; and
replacing the detected computation with an element by element computation between the first matrix and the second matrix in response to determining that the replacement will result in the same output.
7. The method of claim 1 , further comprising extracting channel features from the audio input.
8. The method of claim 7 , further comprising generating the first i-vector further based on the extracted channel features.
9. The method of claim 7 , further comprising: associating the extracted channel features with the generated user account.
10. The method of claim 1 , wherein the metadata elements are any of: phone number, ANI, personal identifier number, IMEI, IMSI, end device make, and end device model.
11. The method of claim 1 , wherein the first i-vector is generated by: the end device, the AIS, or any combination thereof.
12. The method of claim 1 , further comprising:
receiving a second i-vector generated from a second end device, and associated metadata; and
generating a match score between the second i-vector and the first i-vector, in response to matching at least one metadata element value between the metadata of the first i-vector and metadata of the second i-vector.
13. The method of claim 12 , further comprising:
generating a notification that the speaker is authenticated, in response to the match score exceeding a first threshold value.
14. The method of claim 12 , further comprising:
generating a notification that the speaker is not authenticated, in response to the match score being below a first threshold.
15. The method of claim 12 , wherein the second end device is a first end device.
16. The method of claim 1 , further comprising:
receiving a second i-vector generated from a second end device, and associated metadata; and
generating a match score between the second i-vector and each of a plurality of first i-vectors.
17. The method of claim 16 , further comprising:
generating a notification for each match score exceeding a second threshold value, as a possible identification.
18. The method of claim 17 , wherein the notification further includes metadata associated with each of the first i-vectors for which a match exceeded the second threshold.
19. The method of claim 16 , further comprising:
generating a notification based on the first i-vector of the plurality of i-vectors which has the highest match score with the second i-vector.
20. The method of claim 1 , wherein generating a user account is performed in response to determining that the first i-vector is not enrolled.
21. A non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to perform a process, the process comprising:
generating a user account, the user account comprising: a user identifier based on one or more metadata elements associated with an audio input received from an end device;
generating a first i-vector from an audio frame of the audio input, a trained T-matrix, and a Universal Background Model (UBM), wherein the first i-vector generation comprises an optimized computation; and
associating the user account with the first i-vector, and wherein the optimized computation comprises:
generating a feature matrix based on a feature vector;
generating a Gaussian Mixture Model (GMM) mean matrix based on a plurality of mean vectors associated with a plurality of GMM components; and
generating a delta matrix based on the feature matrix and the GMM mean matrix.
22. A system for enrolling a speaker in a speaker authentication and identification system (AIS), comprising:
a processing circuitry; and
a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:
generate a user account, the user account comprising: a user identifier based on one or more metadata elements associated with an audio input received from an end device;
generate a first i-vector from an audio frame of the audio input, a trained T-matrix, and a Universal Background Model (UBM), wherein the first i-vector generation comprises an optimized computation; and
associate the user account with the first i-vector, wherein the system is further configured to:
generate a feature matrix based on a feature vector;
generate a Gaussian Mixture Model (GMM) mean matrix based on a plurality of mean vectors associated with a plurality of GMM components; and
generate a delta matrix based on the feature matrix and the GMM mean matrix.