Computer-implemented detection of anomalous telephone calls, for example detection of interconnect bypass fraud, is disclosed. A telephone call associated with user devices is analyzed remote from the user devices. A first set of multiple features, for example Mel Frequency Cepstral Coefficients, is derived from a call audio stream. The first set is converted to an embedding vector, for example via a model based on a Universal Background Model comprising a Gaussian Mixture Model, which model is preferably configured based on a training plurality of first sets of multiple features derived form a corresponding training plurality of audio streams. Occurrence, or probability of occurrence, of an anomalous telephone call is determined based on the embedding vector, for example via a back-end classifier, such as a Gaussian Backend Model, which classifier is preferably configured based on labels associated with the training plurality of audio streams.

A system and method for authenticating a caller may include receiving an incoming call from the caller, determining a gender of the caller, and selecting, based on the determined gender, to search for the caller in one of: a watchlist of untrustworthy female callers, and a watchlist of untrustworthy male callers.

In various aspects, a system that receives a first message, wherein the first message comprises a communication device identification and a feature tag having a language indication that indicates a language configuration of a user equipment. In response to the receiving the first message, storing the language indication and associating the language indication with the communication device identification. The system determines whether an incoming call is not authenticated and in response to the determining that the incoming call is not authenticated, transmitting a textual message to the communication device using the language configuration to indicate the incoming call is not authenticated.

A system and method for authenticating a caller may include receiving an incoming call from the caller, determining a gender of the caller, and selecting, based on the determined gender, to search for the caller in one of: a watchlist of untrustworthy female callers, and a watchlist of untrustworthy male callers.

Call spoofing can be mitigated by providing geolocation information to the called device. For example, when a call rings, a geolocator can be invoked and the incoming call display screen can show a carrier logo and/or a geolocator globe illustrating the location of the call originator. The geolocation session initiation protocol data can be confirmed by a network device and compared against carrier specific data of the calling device to authenticate voice calls for called devices. In one embodiment location data of the calling device can purposely be shared in order to facilitate the mitigation of call spoofing.

Methods and systems are described for processing calls. An example method may comprise receiving a message for establishing a call. Identification information in the message may be compared to screening data. If a match is found, the message may be forwarded to a screening server. The screening server may establish a call based on the session and provide information indicative of a level of trust associated with the call.

A system and method are disclosed for training a machine-learning model to detect characteristics of fraudulent calls. The machine-learning model is trained using audio clips, voice recognition, call handler feedback and general public knowledge of commercial risks to detect and divert fraudulent calls, thereby alleviating the burdens otherwise placed on call center service representatives.

An example method of operation may include one or more of identifying an outbound call placed by a mobile device subscribed to a protected carrier network, determining the outbound call is destined for a destination telephone number that was stored in a call history of the mobile device, determining the destination telephone number is a scam call suspect telephone number based on one or more identified call filter parameters associated with the destination telephone number, and forwarding a scam call notification to the mobile device while the outbound call is dialing the destination telephone number.

A method of verifying IMEIs and chipset S/Ns of devices within a wireless communication network. The method comprises receiving a request from a device to access the wireless communication network and receiving an international mobile equipment identity (IMEI) and serial number (S/N) from the device, wherein the IMEI and S/N are included on a chipset of the device, and wherein the S/N is the S/N for the chipset. The method further comprises comparing the IMEI and S/N with a database to confirm the authenticity of the IMEI and S/N. Based upon the authenticity of the IMEI and S/N, the request is either granted or not granted for the device to access the wireless communication network.

A server can receive a plurality of records at a databases such that each record is associated with a phone call and includes at least one request generated based on a transcript of the phone call. The server can generate a training dataset based on the plurality of records. The server can further train a binary classification model using the training dataset. Next, the server can receive a live transcript of a phone call in progress. The server can generate at least one live request based on the live transcript using a natural language processing module of the server. The server can provide the at least one live request to the binary classification model as input to generate a prediction. Lastly, the server can transmit the prediction to an entity receiving the phone call in progress. The prediction can cause a transfer of the call to a chatbot.