MHA wants start-ups to create voice recognition software to combat cyber call fraud

In November 2022, the TRAI proposed Calling Name Presentation (CNAP), a feature that aims to allow users to identify callers. [Representative image]

The Ministry of Home Affairs (MHA) is scouting for voice recognition software to rein in the menace of cyber frauds that take place over phone calls.

The home ministry has floated a requirement as part of the Defence Innovation Start-up Challenge, which says that the software will be used to analyse voice samples recorded by victims of cyber fraud.

“The VRS will be able to store voice samples of different individuals of interest that can be uploaded by law enforcement agencies (LEA) and match them with test samples in order to identify suspects,” according to the problem statement issued by the MHA on the DSIC website.

The move comes against the backdrop of a spurt in digital payments and the instances of cyber frauds being carried out through social engineering growing manifold.

“On several occasions, the victim is able to record the voice of fraudsters. The goal of this project is to design and develop a voice recognition system that can assist law enforcement agencies in identifying and tracking individuals who have committed or attempted to commit cyber fraud...” the problem statement said.

How it will work

According to the minsitry's requirements, the proposed system should be able to store voice samples along with additional information that can be imported from another system with the help of an API.

The system should also have the capability of grouping voice samples based on their similarity. “The ability to match individual test samples with the samples available in the database, and display near matches decreasing order of similarity,” said MHA.

It is not clear what database the MHA is referring to hear, or how the voice data will be collected for this particular database, raising concerns about surveillance and infringement of privacy.

“The system should also have the eliminate background noise and other distortions from the voice samples,” it said adding that information gathered from this system should be able to be used as evidence in a court.

Crypto analysis

In other problem statements, MHA floated the requirement for a cryptocurrency intelligence and analysis tool to be able to monitor the usage of cryptocurrencies on the darknet.

“The tool should have the ability to crawl these websites on a regular basis and gather information about the public key addresses being used for payments. The collected information should then be organised into records that consist of the website name, date and time of the transaction.....”, a problem statement on the DSIC website issued by the ministry said.

In another problem statement, MHA has said it wants a solution that can analyse bank statements to assist in detection of cyber crimes.

“The tool will need to be able to handle bank statements in various formats and columns provided by different banks and be able to convert them into a common, structured format,” the MHA said.

“This will enable the tool to analyse transactions and generate intelligence using traditional analytical methods. The tool should also be able to decipher narrations and provide capabilities for link analysis, timeline analysis, and other standard reports and queries,” it added.