It works by measuring "micro-changes" to the frequency of the human voice and relaying to the operator, in real time, the level of risk that the speaker is being deceptive. At the start of the conversation, the software takes the caller's normal voice as the benchmark and accounts for the possibility that changes may be caused by nerves. If the caller is deemed by the operator to be low risk, using the test results to support their own judgement, they are fast-tracked and avoid more rigorous vetting. Those deemed to be at higher risk of lying must supply further evidence to support their claim.
So, is it a step forward in protecting honest tax payers' cash, or a decidedly dodgy attempt to jail the nervous?
ENA3 - spoken language