BATVOX – Forensics Speaker Recognition

BATVOX is an expert 1:1 voice biometric tool designed for forensics experts and scientific police to perform speaker verification and compile expert reports as evidence in court.

BATVOX uses a pioneering voice biometrics technology (text independent, channel independent and language independent) based on unique information extracted from an individual’s vocal tract, thus obtaining a high level of accuracy and reliability.

Key competitive advantages:

– Performance is much higher in real forensic environments than for fingerprints. For some specific tasks, the performance of an automatic system can even be far superior to the performance of human listeners

– High accuracy thanks to Reference Population and Impostors (used for normalization and Likelihood Ratio (LR) calculation)

– Graphic representation of the LR: shows that it is X (=LR) times more likely that the unknown voice recording and the suspect voice model match

– Creation of reports to present results in court

– Proven in the most demanding and challenging environments, BATVOX works with low quality speech signals

– Only 7 seconds of net speech are needed to perform identification, and 40 seconds to create the speaker model


– Allows users to have an additional biometric tool to perform speaker verification thus surpassing and challenging traditional barriers across any telephony channel and any language

– Increases strength of the final result by combining automatic speaker identification with classical phonetic-acoustic method (‘hybrid approach’)

– Certainty of identification within a closed list of suspects to pursue case investigation, providing the precision and reliability required in court hearings

– BATVOX is an expert forensic tool, fast, transparent and easy to use

– A complete solution to give full support to specialists’ investigation work

Categories: , ,