Watermark your Audio with Hidden Text

Back in early 2011 when I started uploading my music to SoundCloud (which was a new platform and was gaining popularity), I wanted some way to add watermark. While album art in .mp3 helped, it was easy to remove. So I started exploring other ways to add hidden elements that could not be – at least without degrading the audio quality – easily removed. I was also too much into exploring encryption, RSA, the enigma machine and the likes of it at that time. For some reason, easter eggs, treasure hunts have always fascinated me.

Coming back to the topic, Image Line's FL Studio came to the rescue with Harmor. The idea was simple:

1. Decide the text or message to add to the audio
2. Find a way to convert this text into audio
3. Use this watermark audio as a layer to merge with the main audio file in any DAW

As easy as it might sound, converting text into audio isn't as intuitive. But mapping colours to frequencies could be. How about black = silence & white = all frequencies?

That's exactly what Harmor allowed. It could import an image file and convert it into a spectrogram. How cool is that?!

So I made a text file in MS Paint (yep, to cut a potato you don't need a sword. A knife does the job) and exported it as an image file. Here's an example:

I've kept the text in white just to show as an example. Feel free to experiment with other colours to see what works best for you. I then imported the image to Harmor, tweak some settings and there it was. An audio file with my name in the high frequency region, all set to be added as a layer in my projects. Notice that while this audio can be heard, it's loudness can be reduced. Besides, most of the audio I was working with at that time had some masking frequencies in the 7-8k range which was enough to make the watermark go unnoticed.

Here are the examples in Acon Digital's Acoustica. The first image is with the spectrogram view, zoomed-in vertically. The frequency values are in the y-axis:

And here is the waveform view, which can be reduced in gain before blending with the target audio:

If you also want to use this technique to watermark your audio, or maybe add secret messages that your listeners might explore in the future, give this a shot. FL Studio has a tutorial on their website with the step by step guide:

If you found this article helpful, let me know in the comments below or using the contact me form. Feel free to share this article with anyone who would find this interesting.