Making Sense Of The Audible WorldSome Animal Biology

A remarkable feature of the human auditory system (and that of many animals) is that the audio processing centre of the brain naturally filters out short-bounce echoes. When we do hear echoes, they tend always to be long-bounce echoes.

Just like microphones, our ears pick up every sonic vibration in the audible range from the environment around us. This includes sound that reaches us via a direct path from its source but also from a very large number of indirect paths, which are short-bounce echoes.

The simplest indirect paths are single and double reflections off walls and objects but there are also many complex multiple reflections (reverberations). If our brains didn’t filter these, we would struggle to make sense of each other and the audible world.

How Our Technology WorksArtificial Echo Modulation

Following many years of research and the accumulation of a large pile of patents, we have solved a seemingly impossible problem. A clue to the solution lies in the little biology lesson above but let’s define the problem first.

The problem is how to embed hidden data (bits) into all kinds of broadcast audio in such a way that the data can be recovered quickly, efficiently and with 100% reliability, but the encoded audio must remain indistinguishable from the original audio.

It’s easy to add bursts of high frequency audio to hide signal data in an original sound track. But that doesn’t work in broadcasting or compressed streaming because broadcasting systems and compression algorithms filter out high (and very low) frequency audio.

Our technology has to work in the audible domain but also must not, itself, be audible. As is so often the case, the inspiration for our basic invention came from nature.

We exploit the fact that the brain naturally filters short-bounce echoes. We take source content containing all kinds of audio (speech, music, sound effects etc.) and we modify it very slightly by introducing artificial echoes.

By modulating these artificial echoes we can encode data bits into the audio content. Each individual audio watermark is an echo sequence that occupies a short section of the audio and carries a fixed number of bits. We call the bit sequence an audio code.

How We Package The TechnologyEncoding & Decoding Made Simple

We’ve developed software and hardware products to implement our encoding algorithms in such a way that it’s very easy for customers to embed individual or sequential audio codes into pre-recorded content and live broadcasts.

We’ve developed a range of SDK products to enable developers to build our audio watermark decoder libraries into any app with minimal effort and with complete creative freedom.

And we’ve developed static decoder libraries for deployment onto tiny, low-cost chips to enable toys and mascots to respond to audio codes in TV shows and at live events.