Ask Vir Ask Vir

Revealed: how the tapes were faked

Last November, Outlook published extracts from the Radia tapes.

These were conversations that Niira Radia, head of India’s largest PR company, whose clients include the Tatas and Reliance, had with several individuals over six months in 2009.

At around the same time, CDs and pen drives containing recordings of these conversations were delivered anonymously to newspapers, magazines and TV channels all over the country.


   These tapes had apparently been made by the income-tax department which had tapped Radia’s phone. Though there were said to be over 5000 conversations, only a tiny fraction were leaked. But the leakers had organised them well, providing transcripts and arranging the tapes so that Radia’s conversations with relatively well-known people were bunched together and featured prominently.


   Though there were at least 25 journalists on the tapes – which you would expect given that Radia’s job was to do PR for her clients – the first set of leaks prominently featured three editors, all of whom duly had their faces plastered on the cover of Outlook.


   One of them was me.


   My conversations with Radia were not about telecom, despite packaging that suggested the contrary. They occurred in the period just after the UPA had won its second term and the Congress and the DMK were squabbling. I spoke to several sources, including Radia, during that time, extracting information about what was happening by probing and stringing them along.


   The problem was that while I did indeed speak to Radia, the tape that was published differed significantly from my recollection of the conversation. Obviously, it had been edited, doctored and manipulated. I immediately wrote out a response on my website stating this. I repeated this in the Hindustan Times and I told everybody who would listen (interviewers, the Public Accounts committee, the Press Council, etc.) that the tapes were not authentic.


   Having made my position clear, I looked for ways to prove that the tapes had been doctored. This turned out to be more difficult than I had anticipated. Audio labs told me that with modern technology editing and manipulation done to digital files by a professional can be almost untraceable.


   Crestfallen, I abandoned this line of investigation.


   Then, several months later, a Shanti Bhushan CD was sent anonymously to the media. This purported to contain a 3-way conversation between Shanti Bhushan, Amar Singh and Mulayam Singh in which Bhushan suggested that his son, Prashant, could fix a highly respected judge.


   When I heard the tape, I knew that it was a fake. I don’t always agree with the Bhushans but their integrity is beyond reproach. Moreover, the judge in question is scrupulously honest and highly respected. Obviously, this tape had been faked or doctored as well.


   The problem was that the voice on the tape sounded – to the naked ear, at least – like Shanti Bhushan’s. Moreover, a government lab said the tape was authentic. Bhushan filed a complaint with the Delhi Police who sent the CD to another government lab. This lab also said that the CD was authentic.


   I understood then why the audio technicians I had spoken to had been so discouraging. With modern technology, it is possible to create near-perfect fakes.


   Then, Prashant Bhushan surprised me. He found two labs, one in Hyderabad and one in America that conclusively demonstrated how the CD had been faked and the conversation fabricated.


   Encouraged by the Bhushan case, I decided to try again. Because I knew that there would always be a question mark over any tests conducted by Indian labs, I looked for top-quality labs abroad, seeking out forensic investigators who knew nothing about the case, had never heard of me, and had no axe to grind.


   I first went to Forensic Audio, a respected Los Angeles company regularly used by the US Secret Service and the FBI. Kent Gibson, Forensic Audio’s founder, told me what I already knew: it might be impossible to find evidence of editing. I asked him to go ahead anyway.


   He used five different techniques, including digital spectral analysis, searching for anomalies using SIS Editracker software and looking for discontinuities in background noise.


   His conclusion both vindicated and relieved me. He found four places where the tape had been edited or manipulated. He declared, “I can say with 90 per cent certainty that the tape cannot be considered authentic.”


   Finally, I had the proof I needed. But because I wanted to be doubly sure, I sent the same tape to Lawdio Inc, another audio lab used by the US Attorney’s Office and law enforcement agencies. Once again, Lance McVickar, President of Lawdio, warned me that phone conversations were notoriously difficult to test for evidence of fakery. But McVikar downloaded the tape from the Outlook website and tested it anyway.


   Forensic Audio had found four instances of doctoring. McVikar found five instances of manipulation with “missing words and broken phrasing”. His conclusion was that “within a reasonable degree of certainty” the tape was not original and had probably been ‘manipulated’.


   With two audio labs in the US confirming that the tapes were not authentic, that left one bit of unfinished business. Though Outlook had not carried it, another magazine had published what purported to be a conversation between Radia and myself in which I seemed to agree to write a column to her specifications.


   I remembered talking to Radia for two columns I was writing about the way in which businessmen were cornering India’s scarce resources. I spoke also to Tony Jesudasan, who handles PR for ADAG and one of the two columns specifically mentioned the conversations, naming both Radia and Jesudasan.


   The problem was that the conversation I actually had with Radia was significantly different from the conversation on the leaked tape.


   Lest I be accused of having both US audio labs in my pocket, I sought out a third audio expert, Paul Baker of Audio Forensic Services in the UK. Baker conducted a different kind of test. He compared a sample of my voice with the voice on the tape. His conclusion was even more shocking. The two voices, he said, “do not match”. He added that having analysed my voice sample and compared it with to the voice on the tape, there were “variations in pitch, pause, pace, inflection and resonance tonal quality”.


   So, I had been Shanti Bhushan-ed. It sounded like me but the tape had probably been electronically synthesised from different conversations. To the naked ear, it sounded like my voice. But a good audio lab could tell the difference.


   With three respected labs in America and England saying that the tapes were manipulated or faked, I don’t think anyone can claim any longer that they are authentic.


   But two questions remain. First, who did it?


   Frankly, I don’t know. Neither does the government – or so it says. Its current position before the courts is that these tapes were not leaked from the IT department and that perhaps service providers had illegally tapped Radia’s phone. Moreover, the government has not been able to convincingly explain why the taps were ordered in the first place. It has changed its story from income-tax investigations to alleged threats to national security.


"Though it is an obvious forgery, the government still maintains that it is genuine and has lab reports to back its case."

   Off the record, government sources told me last year that corrupt officials had made the tapes at the behest of corporate interests and that genuine and false tapes may have been mingled in the IT department’s archives. The tapes were leaked as a diversion, they suggested, to throw the media off the scent while those actually involved in 2G used the time to clean up their money trails before the CBI eventually charge-sheeted them.


   On the other hand, corporate sources told me that this was too major an operation for any business house. They said that the leaks were a symptom of a war within the Cabinet, where ministers were even bugging each other’s offices. I laughed dismissively then. I am not laughing now.


   The second and more important question: what should the media do when presented with seemingly convincing tapes?


   I have no right to get self-righteous about Outlook because, in my career as an editor, I have also carried tapes without verification. Besides, how reliable is verification, anyway? Consider the Bhushan tape. Though it is an obvious forgery, the government still maintains that it is genuine and has lab reports to back its case.


   So, here’s what I think. Technology has now advanced to the stage where fake documents, morphed photos and fabricated tapes are easy to create. As journalists, we lack the expertise to tell what is genuine and what is a fake. So, perhaps the time has come for us to be more cautious. Just because something sounds right, it does not follow that it is genuine.


   Technology can make fools – and liars – of us all.



Note on Authenticity Tests


Three audio laboratories in the US and England were contracted to run tests on the audiotapes of Vir Sanghvi talking to Niira Radia.


   For purpose of authenticity, the labs were asked to use the tape recordings available on the websites of Outlook and Open Magazine.


   This is what they found.


   The first tape analyzed was of a conversation in which Vir Sanghvi allegedly discussed passing on messages to Ahmed Patel and mentioned dealing with Rahul Gandhi.


   The tape was first sent to Forensic Audio, a respected Los Angeles company regularly used by the US Secret Service, the FBI and Los Angeles Courts.


   Forensic Audio found that the tape had been doctored with at least four instances of editing.


   Its founder and head, Kent Gibson wrote, “I conclude with an estimated 90 per cent certainty that the recording has been modified and cannot be considered authentic”.


   The same tape was then examined by Lawdio Inc., a top audio lab that is widely used by the US Attorney’s office and local police forces.


   The President of Lawdio, Lance McVickar downloaded the conversation from the Outlook website.


   He found five places with “missing words and broken phrasing.”


   He declared: “I have come to the conclusion that this audio most likely was manipulated.”


   A second recording purported to contain a conversation between Niira Radia and Vir Sanghvi in which Vir Sanghvi appeared to agree to stage an interview with Mrs Radia’s client Mukesh Ambani and offered to write a column based on her briefing.


   Audio Forensic Service a respected British audio lab run by Paul Baker was asked to examine this recording, taken from the Open magazine website.


   Mr Baker asked Vir Sanghvi to record the exact same words as the man in the recording to compare the voices.


   His conclusion was that the voices did not match. Vir Sanghvi’s voice (as revealed in a spectrogram) was not the voice in the recording.


   He concluded that the two voices “appear to be completely different and do not match.”


   He added “Having analysed the two files on an auditory level with various tools, they show variations in pitch, pause, pace, inflection and resonance tonal quality.”


   Therefore even this conversation was fabricated.


   It is possible that words from different conversations were digitally strung together to create a manufactured conversation. To the naked ear, this might sound like a smooth conversation. But under scientific analysis, its fraudulent origins reveal themselves.


   All three reports follow:


I, Kent Gibson, state the following, of which I have personal knowledge:


I am the founder of Forensic Audio (, which is a 15 year old company based in Los Angeles, California. Regular clients include the US Secret Service, the FBI, LA Superior Court, LA County Sheriff, LA Public Defender's Office, Pasadena PD Homicide Assaults, Santa Clara Sheriff's Dept, Santa Rosa County, San Bernardino County Sheriff's Department and many private law offices and various other courtroom representatives. I am a Certified Audio Forensic Examiner for LA Superior Court, and chosen by the LA County Sheriff as a contract examiner for the county. . Kent Gibson has a BA from Yale University in Psychology of Communication and a Specialty in Linguistics.  He has an MA from Stanford University Department of Communication specializing in audio and video.
 specializes in examining and preparing audio and visual evidence for use during litigation. Additionally Forensic Audio authenticates recordings looking for alterations and edits, performs voice identification and prepares certified transcripts


            In the present case, I was contracted by Raaj Sanghvi to authenticate a recording. The techniques I used are the following:




1. Critical listening using very high quality equipment, speakers and


headphones. (Identification of suspect pops and jumps in tonality, gaps in the recording, etc.)


2. Examination of syntax and vocal context looking for non-sequitur transitions, truncated words or breaths and illogical juxtapositions.  


3. Digital Spectral analysis using spectrograms.


4. Search for phasing anomalies in stable technical signals – using the SIS


Editracker Software.


5. Search for discontinuities in background noise parameters – using the SIS


Editracker Software.





In making a judgment of authenticity, I rely on at least three or more of the above measures to confirm an edit. In this case, I was able to confirm FOUR instances of editing by THREE OR MORE of the above measures.  




Kent Gibson - Forensic Audio & Video Examiner

3251 Oakley Drive, LA CA 90068


CONFIDENTIALITY NOTICE:  This e-mail transmission, and any documents, files or previous e-mail messages attached to it may contain confidential information that is legally privileged.  If you are not the intended recipient, or a person responsible for delivering it to the intended recipient, you are hereby notified that any disclosure, copying, distribution or use of any of the information contained in or attached to this transmission is STRICTLY PROHIBITED.  If you have received this transmission in error, please immediately notify the sender.  Please destroy the original transmission and its attachments without reading or saving in any manner.  Thank you.  The contents of this email are confidential and protected from disclosure by the attorney work product doctrine.


Lawdio inc.




It is my opinion within a reasonable degree of certainty after running some audio authentication tests, the digital recording which was publicly available and posted on the Outlook Magazine website is not original.

After critical listening via headphones and visually inspecting the wave form of the file using my digital audio workstation of the mp3 file provided I found five places in the file where an edit or drop out of audio occurred.  It is not possible to say with 100 percent certainty that the file was in fact edited as today’s phones glitch via interference etc and editing can be done to digital files by a professional and can be almost untraceable, but the five spots seem to have missing words and broken phrasing, therefore I have come to the conclusion that this audio most likely was manipulated.

Lance McVickar



Lance McVickar, President / Head Forensic Audio / Video Engineer.  Lance McVickar has had extensive experience working in conjunction with city, state and federal law enforcement agencies, and the United States Attorney’s Office, and has been a forensic audio engineer for a decade.  He has been an audio engineer for over twenty years and is expert in all types of analog and digital forensic audio and video technologies. He is also the main contact for client relations.  





  • Ritesh Jain 26 Jan 2012

    Please restart Counterpoint. I am missing it

  • prabhat kumar 20 Dec 2011

    i didn't believed those allegations from the very moment, they were publihed.
    please continue your HT column.

  • hemant 13 Dec 2011

    Vir bhiya, it’s a big relief for all of Vir fans, we all waiting for encounter in HT. .

  • To view all please click on More Comments below
More Comments:(14)Posted On: 29 Oct 2011 12:04 PM
Your email id will not be published.
Security code:
Captcha Enter the code shown above:
Your email id will not be published.
Friend's Name:
Friend's E-mail:
Your email id will not be published.
The Message text:
This email was created by [your name] who thought you would be interested in the following Article:

A Vir Sanghvi Article Information

The Vir Sanghvi also contains hundreds of articles.

Additional Text:
Security code:
Captcha Enter the code shown above:

CommentsOther Articles

See All

Ask VirRead all

Connect with Virtwitter

@virsanghvi on
Vir Sanghvi