Specifications

FIELDMARSHAL INTEGRATED SECURITY SERVICES
24
Suspect Detection System
The TMP-SDS core technology is based on proprietary software; it is an “expert
system” that emulates an investigator’s Modus Operandi (MO) by incorporating
artificial intelligence techniques and algorithms. All hardware elements are best-
of-breed off-the-shelf third-party components. The technical solution is comprised
of a front-end, the ‘Test Station’, and a back-office where multiple-station and
multiple-site data is stored, managed and distributed.
"TMP-SDS" is a fully automated system not requiring human control and/or
operation.
All of the TMP-SDS products are an all in one unit which includes BIOMETRIC
and Documentation Verification features in addition to the interrogation
capabilities. The TMP-SDS technology enables conducting hundred of
interrogations without help of professional interrogator interviewers. In 5 minutes
test the system can identify terrorists, employees who has hostile intents,
criminals, smugglers or collaborators and direct further interrogation. The
operation of the system is simple and requires only 2 hours training of any soldier
with basic computer knowledge.
The TMP-SDS 1002 Test Station serves as the point-of-contact with the individual
being examined. The Test Station features an ergonomic design and is adapted
to the environmental conditions (limited space, necessity to queue, strong
background nois etc.) common to border control points.
The TMP-SDS 1003 is an “Internal Threat” prevention and detection system
which was developed for pre employment and repetitive employees screening. It
is also fully automated system and require no involvement of professional
interrogators or interviewers.
The TMP-SDS 4M is a military grade Rapid Intelligence Interrogation Solution.
This new mobile system is already being used by intelligence units in Israel. The
system enable military units to efficiently interrogate hostile or indifferent civil
population. Using the TMP-SDS 4M will enable military unit which might have no
knowledge of the local spoken language to identify who among the civilian is a
wanted terrorist, where there might be nearby ambush or mine trap and above all
– save lives.
FreeSpeech - Biometric speaker
verification system
Independent of conversation content, the language spoken and accent,
FreeSpeech transparently verifies customers within seconds. Enrollments take
place in the background of natural conversations between the customer and
contact center agent. By eliminating cumbersome authentication processes,
FreeSpeech enhances the customer experience, reducing call duration and
associated costs, and allows agents to focus on providing quality service.
FreeSpeech 6.5 enables real-time operational and transactional risk management
by biometrically analyzing callers’ voices. Fraudsters watch lists can be used to
identify fraudulent activity.
Main Features
ALGORITHMS
• Language and accent independent
• Text independent
• State-of-the art accuracy
• Verification and identification capabilities
• Voiceprint quality assurance via consistency and DTMF/dial-tone detection
• Supervised and unsupervised voiceprint adaptation
• Multiple threshold/tuning settings
INTERFACE
• Simple and easy-to-use Web Service APIs enable secure, prompt and
platform-independent integration
• Easy-to-use SDK
• Identity management via LDAP
• Application data - easily attached to any system entity (i.e., speaker,
voiceprint)
• Authentication feedback capabilities - enabling controlled performance
evaluations
• GUI embedded in Windows Explorer, enabling offline system operation using
prerecorded audio files
• Scripting tool for management and performance evaluations