The system delivers speaker identification in as little as seven seconds and compares the speaker(s) to a watchlist of people of interest in real-time, while a conversation is in progress, or as part of a batch process completed after the conversation has ended.
The system can process both mono and stereo audio inputs as it is capable of splitting the caller´s voice from the called party’s voice with extreme precision. There is no need to embark on costly, time consuming projects to modify your equipment; our system works with both mono and stereo.
As the system utilizes technology which is intrinsically language independent, it can be effectively used to identify speakers of any language without the need for adaptation, changes or reconfiguration.
CONTINUOUS VOICE IDENTIFICATION
Provides investigators not only verification of whom is speaking, but also the identity of the inmate(s). Prints can be extracted from any source and used to match in any other channel, regardless of language or text. Voices on NON-INMATES can also be run against the database.
CREATE VOICE PRINTS ON REQUEST
Prints are automatically created and continuously improved, delivering the highest accuracy possible. Prints can also be searched through historical calls, allowing investigators to listen to and extract potential intelligence.
IDENTIFY PIN IMPOSTORS
Identify PIN sharing, as well as the identity of the PIN thief, allowing investigators to focus their efforts. In the event that ECHO cannot positively identify an impostor, the system will still send notification that the speaker was not the owner of the PIN.