Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now


CSI Cambridge

This week Phil Ward look at the fundamentals of audio forensics with CEDAR Audio’s managing director Gordon Reid

Cambridgeshire-based CEDAR Audio is renowned for developing some of the most sophisticated digital audio analysis and processing available today. But as well as attracting discerning clients in recording, broadcast, mastering and post-production, the company has just announced the CEDAR Forensic Audio Starter System – a product that, and we quote, “provides an attractive price point for police forces and security agencies”. Hello? Something for Sergeant Pepper, perhaps? Silence in court: CEDAR Audio Managing Director Gordon Reid has all the admissable evidence collected from the scene… How did CEDAR Audio first get into forensic audio investigation? “I was avoiding the English winter by working in Sydney in November 1993 when my Australian dealer was approached by the Australian Federal Police. Having reassured him that their visit related only to signal processing, they told him that their audio lab was interested in the possible forensic applications of audio restoration. I didn’t feel that our existing technology was suitable, but they had found that a combination of existing filtering techniques and CEDAR’s noise reduction could increase intelligibility and make recordings considerably more listenable. So, after a year of discussion and testing, I again avoided the English winter by returning to Australia in December 1994 to install the first CEDAR system in a forensic lab.” Where did it go from there? “Our next forensic installation was in Wellington in May 1995, which turned out to be a serious meteorological mistake on my part. This time, it was a combination of adaptive filtering and CEDAR that had interested the New Zealand Police, and they obtained some remarkable results processing a recording made in a very noisy military vehicle that I can’t tell you about. Consequently, we decided to develop a dedicated CEDAR forensic system that combined existing technology, adaptive filtering and some new ideas that we had.” So that’s what you did? “No. We found that the hardware of the time was unable to support the system that we wanted to develop, so we shelved the idea for a few months… that eventually lasted for seven years. It wasn’t until we launched CEDAR Cambridge that we could implement the adaptive filters and other processes that we wanted to offer, and the first CEDAR Cambridge Forensic System was announced in October 2003.” So anyone with the audio skills can now get into audio forensics? “Umm… no. The forensics and security arenas are highly regulated and practitioners generally have to be able to present long CVs of work with police forces and agencies that have no addresses and only three or four letters in their names. It’s a bit Catch-22: you need to be recognized to get the jobs, but you need to do the jobs to become recognized. Perhaps the best way into the industry is to work for one of the recognized labs, such as the one at the Metropolitan Police where our Forensic Manager earned his stripes, so to speak.” But audio forensics can now get the wanted information out of most recordings? “Not at all. We still hear recordings made using a surveillance recorder buried at the bottom of a handbag underneath a thick jacket and a woolly scarf, tucked away under a heavy table in a noisy restaurant, and I think that I can say with confidence that signal processing – no matter how advanced – is not the answer.”