Audio in Surveillance: Worth Another Listen

Poor-quality audio encoders built into CCTV cameras mean that end users could be missing out on the potential to capture vital evidential information.

Click here to view Figure 1.

According to Paul Murphy, marketing manager at IndigoVision, audio hasn’t been seen as a priority in surveillance, and although many of the latest IP cameras boast of high-definition image capture, they fall short of the mark when it comes to the quality of their audio offering. He told IFSEC Global:

Audio is often seen as a niche application in security, with most cameras using poor audio encoders. If the right audio codec is used — it can be a huge compliment to video evidence.

Like video, audio needs to be compressed in order to make it easier to store and transmit over network and there are a number of codecs that will compress audio. Most cameras on the market with an in-built audio feature use the G.711 and G.728 compression codecs. These were built originally for the telecommunications industry to transport the human voice over long distances. G.711 was first released in 1971 and G.728 in 1992 — so they have been around awhile.

Murphy continued:

Because these codecs were built around the human voice, they remove all other noise that isn’t in the human voice range. You can hear how a G.711 or G.728 codec affects noises outside the human vocal range when you are placed on hold and have the pleasure of hearing music that is played to you to keep you on the line! You don’t get any bass or high notes.

IndigoVision’s new Enhanced Standard Definition (SD) range (including the 9000 Series) of cameras now all come with a built-in audio features based on the AAC codec, which was developed in the late 90s for digital audio and is the standard format for most music players.

IndigoVision says that it has seen a number of its customers, particularly in the Latin American region, use the IndigoVision solution to provide “help points” around the city to enable people to alert central monitoring to a problem.

Of course, the use of audio in addition to images treads a fine line between security and a breach of privacy. The Information Commissioner’s Office (ICO) Code of Practice suggests on its website that an audio function should be used in a limited range of applications — for example, as an alarm or alert to certain sounds such as breaking glass, or at a help point or as trigger to begin recording in response to a specific threat.

End users have the option of having the mic built-in or connecting to an external mic. Mark Staples at UTC Fire & Security said:

Audio will probably always be better with an external device. But in terms of quality it has certainly come on and is better now than it was five years ago.

UTC’s UltraView 650TVL WDR box camera offers an in-built solution and also provides 650TVL high resolution and 0.01 lux sensitivity.

Trevor Wallace, managing director of UK security systems installer, Metro Security told us that while audio as a built-in feature of a surveillance cameras is a more discreet option — the sound you get will often be compromised:

In nine times out of ten we prefer to use separate, more advanced conferencing microphones to ensure, for instance, that the conversations by staff in a retail shop which we want to hear are not compromised by other background noises and this is best achieved using higher quality bespoke audio equipment. Built-in mics are certainly improving, but as part of a camera that’s designed to capture images, rather than audio, they still have their limitations.

Of course the type of solution you opt for will depend on the application and high quality in-built audio comes at a price. For Murphy this is well worth paying:

The benefits far outweigh the negatives of high-priced cameras. With AAC you can pick up background noises such as breaking glass, footsteps or gun shots. Having audio clarity helps gives greater evidence for audio applications in security.

What’s your audio solution for video surveillance? Is it something you even need?