Cedar Audio has announced the FNR broadband noise reduction and speech enhancement module for Cedar Cambridge. The module is aimed at audio postproduction and broadcast, as well as for audio forensics and security.

Based on a new noise reduction algorithm, this has been designed to clean up speech recordings suffering from very poor signal/noise ratios, and is capable of performance that would have seemed impossible just a few years ago.

‘I have been enhancing forensic audio material for more than 30 years, and one scenario has proved to be stubbornly problematic, and often impossible to solve,’ says Alan French, who heads Cedar’s forensic and security division. ‘This is the problem of revealing and increasing the intelligibility of voices recorded in crowded cafés and similar environments. We in the forensic audio community encounter this time and again and, while the venues and scenarios may vary greatly from case to case, they all share a common difficulty; how do you reveal a wanted speaker when the signal to noise ratio is very poor, or if the unwanted sound on the recording is actually other human voices?

‘Until now, the best tools available have been Cedar’s single-channel adaptive and lattice filters as well as the NR4 and NR5 broadband noise reduction modules, but FNR is going to be a game-changer for this type of work. Because it’s adaptive, it reacts – and reacts quickly – to changes in the background noise in café-type recordings, and it doesn’t have anything like as much effect on the wanted voices as previous filtering systems. I have been testing FNR for more than six months prior to its release, and it has yielded excellent results on genuine casework. It also works very well on recordings made in moving vehicles because, as in the café scenario, the background noises (primarily, the engine and road noise) are constantly changing, and FNR tracks these changes better than anything else I’ve yet heard.’

More: www.cedaraudio.com