Research  
  Audio Watermarking
< prev next >
 
     
 
 

An audio watermark is a special “code” embedded within an audio signal, which is not easy to detect by human ears, but easy for a computer to extract.  Robust watermarking systems are “perceptually transparent” and robust against transmission, compression, or attacks on analog and digital channels.

This data contained in the watermark can be used to carry any type of digital information such as copyright data, images, lyrics, closed caption, etc.

Base 10 Labs’ watermarking technology makes use of spread spectrum theory to create a robust watermark signal, and a custom psychoacoustic model of human hearing to embed the watermark in the audio stream, making it imperceptible to the average listener. 

The watermark creation and detection are real-time processes, facilitating applications such as real-time close captioning, multimedia data tagging, playback analytics, copyright information, or any on-the-fly data that the user desires to embed in the audio signal.