Voice Risk Analysis (VRA)
Voice risk analysis is a real time system that combines the measurement of physiological levels of voice stress with behavioural analysis and conversation management techniques to enable the detection of truthful statements. It is a proven technique used in the private sector, for example, in better assessing the risk associated with insurance claims.
The objective of this project is to see whether the same techniques can be used successfully in Housing Benefit administration in particular and in welfare benefits in general.
The project is examining the potential of VRA to simplify new claims and review processes, and produce intelligence led risk assessments that lead to more effective, targeted counter fraud interventions.
VRA is currently being tested at a number of local authorities and early indications are that, in addition to identifying the relatively low number of potentially fraudulent claims, it has also been successful at identifying customer error.
The specific objectives of the project are:
- To improve customer service by providing processes that are quicker and less intrusive to the vast majority of claimants
- To provide a service that is more efficient and therefore less costly
- To improve fraud and error prevention and detection
We are planning a second phase of pilots aimed at establishing the potential efficiencies and savings to be achieved from using this technology. This phase will extend to a number of new sites and, in testing different types of implementation, will offer the scope for shared working among smaller LAs.
If the pilot is successful we will consider the case for changes to verification procedures for cases judged to be low risk, potentially reducing the need to issue and process forms and undertake unnecessary and expensive visits. Where cases are judged high risk, counter fraud activity will be undertaken promptly ensuring more fraud is detected and detected earlier, reducing overpayments.
We also expect there to be a natural deterrent effect that will both encourage the self-correcting of claims when faced with follow-up action and, in the longer term, deter fraudulent claims in the first instance.
Once the pilot has been evaluated, we will publish the findings and any recommendations that can be made.