- DC Area
- Jersey Area
- Los Angeles
- New York
- SF Bay Area
- Earlier editions
- India News
- US News
The poor sound quality of the latest taped message attributed to Al Qaeda chief Osama bin Laden has created problems for American analysts in distinctly verifying his identity, senior Bush administration officials said.
Analysts at the Central Intelligence Agency and National Security Agency are facing problems in analysing the tapes, the officials said on Thursday.
The tape had been played on telephone to the Arab satellite television station Al-Jazeera on Tuesday. The high and low frequencies that help distinguish voices may have been wiped out during telephony, a media report said.
"But even if the quality had been superb, government analysts would have had problems trying to verify the identity of the voice," The New York Times reported.
Since the vocal tract physically differs from one person to another, the identity can be established by voice analysis. But unlike verifying someone's identity through fingerprints, voice recognition is indirect.
The science of using computers and linguists to identify individuals by their speech has improved dramatically in the last few years, but it still involves considerable speculation, say experts.
The government's assessment so far that it cannot be absolutely certain that the audiotape broadcast on Tuesday has Osama bin Laden's voice does not surprise voice authentication experts, the newspaper said. "If an impostor wanted to put together a tape, he could fool a lot of people," one expert was quoted as saying.
Experts said the most successful impostor would be a member of Laden's immediate family, though mimicry is often detected by the new technology used in voice authentication.
Government analysts, the paper said, would compile a set of recordings that Laden is known to have made. Most of these recordings came from videotapes in which it is clear that Laden was the speaker. This earlier material provided his speech and varied background noises with which the new tape could be compared.
With the previous recordings as a template, the government analysts would have then fed the old and new tapes into a computer using software that was designed for the intelligence agencies. The programme, The New York Times said, converts speech into bits to compute the degree to which the two sets match.
The resulting score means there is 90 per cent probability of the voice in the new tapes belonging to Laden. By comparison, fingerprinting by law enforcement officials is almost 99.99 per cent accurate.
Back to top
Tell us what you think of this report