SIFT is a stand-alone Speaker Identification Flexible Toolkit, ready to be installed in a portable computer. SIFT has been designed based on the precise and ultra-fast fourth generation of AGNITIO’s Voice Biometrics technology. It allows organizations to apply the power of voice recognition to support multiple law enforcement, forensic and intelligence operations.
SIFT has strong front-end capabilities, such as audio validation and Automatic Voice Activity Detector (AVAD) to remove artifacts, noises and non-voice events. It allows processing and storing of voices and useful information about the speaker. SIFT has been designed to store up to 1,000 different speakers in its database. It even allows the processing of audio files with two speakers in order to identify voices of individual speakers. SIFT also provides the ability to run identifications between unknown voices and speakers. With its comprehensive Voice Biometrics capabilities and flexibility, SIFT is the preferred choice for multiple high precision speaker identification tasks in law enforcement and military organizations, where large parallel processing or integration is not required.