Controlled-environment communication systems are increasingly using voice over internet protocol (VoIP) to serve their users. VoIP allows voice to be sent in packetized form, where audio is encoded using one of several codecs. Because of bandwidth constraints, particularly during peak call times, codecs may be used which sacrifice audio quality for bandwidth efficiency. As a result, several features of communication systems, including critical security features. The present disclosure provides details for systems and methods by which a controlled-environment communication system may shift between codecs to perform security-related features or to alleviate bandwidth considerations. This involves the special formatting of control-signaling messages, including session initiation protocol (SIP) and session description protocol (SDP) messaging.

Techniques described herein include receiving, authorizing, and processing voice commands to control computing devices and perform various device capabilities. In some examples, a user computing device may implement voice command functionality using multiple independent components, with shared security credentials established between different combinations of components. An intermediate voice authorization component may receive and compare voice input data received from a user interface component with voice template data stored securely by a voice data component, to protect against a component becoming compromised by malware or exposure to an untrusted system. Voice commands may be used to execute, disable, or enable various capabilities on the user device, including different device applications and features, and may be authorized by different users with various security and authorization techniques.

In a method and apparatus for improved entity extraction in an audio of a conversation or a call, the method includes generating, at a server, from speech data of a conversation between at least two persons, text data and associated preliminary entity prediction data, using an automated speech recognition (ASR) engine comprising one or more neural networks trained via multi-task training. The method further includes identifying, using the text data and associated preliminary entity prediction data, at least one named entity in said speech data.

A server transmits a task to client devices using push-to-talk (PTT). The server receives a response accepting the task from a device of the client devices. The server assigns the task to a user of the device. The server transmits an indication that the task was assigned to the user of the device to at least a subset of the client devices.

A computer-implemented method for providing an objective evaluation to a customer service representative regarding his performance during an interaction with a customer may include receiving a digitized data stream corresponding to a spoken conversation between a customer and a representative; converting the data stream to a text stream; generating a representative transcript that includes the words from the text stream that are spoken by the representative; comparing the representative transcript with a plurality of positive words and a plurality of negative words; and generating a score that varies according to the occurrence of each word spoken by the representative that matches one of the positive words, and/or the occurrence of each word spoken by the representative that matches one of the negative words. Tone of voice, as well as response time, during the interaction may also be monitored and analyzed to adjust the score, or generate a separate score.

To filter unwanted sounds from a conference call, a voice profile of a first user is generated based on a first voice signal captured by a media device during a first conference call. The voice profile may be generated by identifying a base frequency of the first voice signal and determining a plurality of voice characteristics, such as pitch, intonation, accent, loudness, and speech rate. These data may be stored in association with the first user. During a second conference call, a second voice signal captured by the media device is analyzed to determine, based on the voice profile of the first user, whether the second voice signal includes the voice of a second user. If so, the second voice signal is prevented from being transmitted into the conference call. A voice profile of the second user may be generated from the second voice signal for future use.

Aspects of the disclosure relate to voiceprint tracking and anomaly detection. A computing platform may detect voice information from a call management system. The computing platform may establish voiceprints for employees and clients of an enterprise. The computing platform may detect a call between an employee and a caller attempting to access a client account. The computing platform may identify a first voiceprint corresponding to the employee and a second voiceprint corresponding to the caller. The computing platform may compare the second voiceprint to a known voiceprint corresponding to the client. Based on the comparison of the second voiceprint to the known voiceprint, the computing platform may determine that the second voiceprint does not match the known voiceprint. The computing platform may identify that the second voiceprint corresponds to another employee of the enterprise, and may send a security notification indicating potential unauthorized account access to an enterprise computing device.

Aspects of the disclosure relate to voiceprint tracking and anomaly detection. A computing platform may detect voice information from a call management system. The computing platform may establish voiceprints for employees and clients of an enterprise. The computing platform may detect a call between an employee and a caller attempting to access a client account. The computing platform may identify a first voiceprint corresponding to the employee and a second voiceprint corresponding to the caller. The computing platform may compare the second voiceprint to a known voiceprint corresponding to the client. Based on the comparison of the second voiceprint to the known voiceprint, the computing platform may determine that the second voiceprint does not match the known voiceprint. The computing platform may identify that the second voiceprint corresponds to another employee of the enterprise, and may send a security notification indicating potential unauthorized account access to an enterprise computing device.

Controlled-environment communication systems are increasingly using voice over internet protocol (VoIP) to serve their users. VoIP allows voice to be sent in packetized form, where audio is encoded using one of several codecs. Because of bandwidth constraints, particularly during peak call times, codecs may be used which sacrifice audio quality for bandwidth efficiency. As a result, several features of communication systems, including critical security features. The present disclosure provides details for systems and methods by which a controlled-environment communication system may shift between codecs to perform security-related features or to alleviate bandwidth considerations. This involves the special formatting of control-signaling messages, including session initiation protocol (SIP) and session description protocol (SDP) messaging.

A system is described herein that facilitates the monitoring of inmate communications. The system provides a remotely-accessible means for a reviewer to monitor a call between an inmate and another person. The system includes a monitoring server and a monitoring station. The monitoring server is configured to receive a call and call information from a communication center and process the call for monitoring, schedule a review of the call; and store the call, the call information, and scheduling data. The monitoring station is configured to receive the call and the call information from the monitoring server based on the scheduling data, and to display the identifying information and facilitate the review of the call.