Forensic expert who solved mystery of first 'human bomb' talks on JNU videos

Bengaluru, Feb 28: "I cannot make out anything from the videos in the present condition. However, video forensic technologies, like voice spectrography could easily unravel the truth whether the anti-India sloganeering videos shot at JNU were doctored or not."

This was expressed in an exclusive interview with OneIndia by Padma Bhushan Professor Dr P Chandra Sekharan who pieced together dismembered body of Dhanu to tell the world that it was 'human bomb' which was used to assassin former Prime Minister Rajeev Gandhi

Padma Bhushan Professor Dr P Chandra Sekharan

As many many as eight JNU videos allegedly doctored have now sent to Central Forensic Sciences Laboratory, New Delhi and these are stinging evidence in the JNU row in which students are charged with sedition charges.

Chandra Sekharan, who has enviable experience in forensic sciences, opines videos alone cannot make a strong case for sedition and police has lot work to do beyond video evidence to prove the charges.

How the video forensic is done?

The video and audio forensic has a technology to analyze this area, to see whether the the video was animated, cut and mixed- some portion of video has been removed or some portion added, Chandra Sekharan said

This will also shed light on, is video strip continues or cut and mixed or added. The forensic experts will also keenly work on mimicking aspect. "The experts will work on to know has any individual mimicked thus dubbing the audio strip." Sekharan pointed.

A technology, voice spectrography would be used to determine the doctoring of the audio. "The soundtrack of the footage can be analyzed easily through voice spectrography. The technology is capable of determining whether the voice was original or mimicked one," professor informed.

There are good forensic experts now in India who can table the result on the JNU videos in a span of four days. Chandra Sekharan keenly continued and said, the voice spectrography also finds, does the audio was superimposed from another video.

The mimicked voice or superimposed audio, is generally louder than the natural voice and this a key to differentiate or segregate the two from the footage. As a primitive step the video will be run in slow motion to identify accused persons, later other experiments will come to play.

The synchronization between lip movement and audio will also be studied by the experts in the process. "Sometimes, if dubbing fails to match the lip movement then long shots will also be used by the perpetrators and this angle will also be pursued by the forensic experts." Chandra Sekharan observed.

The experts also have to collect the original voice samples of the accused in the sedition case. The original videos of speeches of Kanhaiya Kumar and others shot in other occasions would be gathered. Or else, they would make accused to read out loud the scripts of anti-India sloganeering heard in the videos in question.

Finally, the lab would compare the original voice samples and voices in the videos to file the conclusion report. Though reconstructing slogans emanating from the mob is hard, the experts can finish the discourse in a span of 3-4 days.

How crucial video evidence in sedition case?

Chandra Sekharan says since the whole case is based on video evidences, the videos play crucial element in the case. However, it is not the only aspect through which sedition charge can be proved.

The investigators also have to look beyond videos too in order strengthen the case. The police has to establish, mainly has any constructive anti-India agenda was being carried out by the students in the campus or not. "Yes videos play crucial evidence, but other aspects too have be investigated."

Chandra Sekharan said if videos were doctored then there won't be any case at all. if not, " I won't be surprised if the court holds the accused guilty as sometimes court wants to create a fear psychosis in the society to stop such activities further," he viewed.

OneIndia News


Please Wait while comments are loading...