Systems and methods for detecting call provenance from call audio
First Claim
1. A method comprising:
- receiving, for a first call between a plurality of parties at a corresponding plurality of devices, data, received at a first device of the plurality of devices during the first call, representing audio from one or more other devices of the plurality of devices;
determining, by a computing device, based on the data representing the audio received at the first device, at least one of a noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at the first device; and
at least one of identifying and verifying uniquely, based on the at least one of the noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at the first device, at least a portion of a path of transmission of the first call between the plurality of devices, wherein the at least one of identifying and verifying uniquely is not dependent on metadata associated with the first call.
1 Assignment
0 Petitions
Accused Products
Abstract
Various embodiments of the invention are detection systems and methods for detecting call provenance based on call audio. An exemplary embodiment of the detection system can comprise a characterization unit, a labeling unit, and an identification unit. The characterization unit can extract various characteristics of networks through which a call traversed, based on call audio. The labeling unit can be trained on prior call data and can identify one or more codecs used to encode the call, based on the call audio. The identification unit can utilize the characteristics of traversed networks and the identified codecs, and based on this information, the identification unit can provide a provenance fingerprint for the call. Based on the call provenance fingerprint, the detection system can identify, verify, or provide forensic information about a call audio source.
-
Citations
47 Claims
-
1. A method comprising:
-
receiving, for a first call between a plurality of parties at a corresponding plurality of devices, data, received at a first device of the plurality of devices during the first call, representing audio from one or more other devices of the plurality of devices; determining, by a computing device, based on the data representing the audio received at the first device, at least one of a noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at the first device; and at least one of identifying and verifying uniquely, based on the at least one of the noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at the first device, at least a portion of a path of transmission of the first call between the plurality of devices, wherein the at least one of identifying and verifying uniquely is not dependent on metadata associated with the first call. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
-
25. A system comprising:
-
a processor; a memory operatively coupled to the processor; a characterization unit for extracting, independent of metadata associated with a first call, one or more characteristics from call audio received at only a single end of the first call, wherein the one or more characteristics include at least one of a noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at only the single end of the first call and storing the extracted characteristics in the memory; a training unit for learning application of one or more of labels to a plurality of prior call audios, based at least partially on characteristics detected from audio from prior calls; and a prediction unit for predicting, with the processor, labels for the first call based on the learned application of the one or more labels from the training unit to the call audio of the first call. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38)
-
-
39. A method comprising:
-
detecting, independent of voice analysis, one or more characteristics associated with a first call solely from data representing audio received at only a single end of the first call, wherein the one or more characteristics include at least one of a noise profile, packet loss characteristics, and speaker-independent artifacts associated with the audio received at only the single end of the first call; determining, by a processor, a first provenance fingerprint for the first call based on the detected call characteristics; and identifying and verifying uniquely one or more devices or networks spanning at least a portion of a transmission path of the first call. - View Dependent Claims (40, 41, 42, 43, 44, 45, 46, 47)
-
Specification