Investigators concerned over privacy as analytical technology recreates MD-11F crash audio

While original audio is protected by law, computational techniques are capable of reconstructing close approximation.

US investigators are trying to address a privacy concern arising from detailed information released as part of the inquiry into the fatal UPS Boeing MD-11F crash in Louisville last year.

The National Transportation Safety Board held a two-day hearing into the accident on 19-20 May, and placed a docket into the public domain containing relevant data.

But it believes some of the data has been used to create simulations of the accident that include audio from the cockpit.

US federal laws prohibit the release of cockpit audio recordings, owing to the “highly-sensitive nature” of verbal communication, says the safety board.

“The NTSB takes these privacy restrictions seriously,” it adds. Only relevant written transcripts are provided.

But while original audio is not released, modern technology is capable of replicating the cockpit environment from other data made available by investigators.

“The NTSB is aware that advances in image recognition and computational methods have enabled individuals to reconstruct approximations of cockpit-voice recorder audio from sound spectrum imagery,” says the safety board.

It specifically mentions data released during the inquiry into the MD-11F accident on 4 November last year which resulted in 15 fatalities including the three crew members of the aircraft.

As a result, the safety board has temporarily taken its public docket system offline while it explores the “scope of the issue” and looks to “evaluate solutions”.