Recording artists welcome new efforts to combat music piracy

Watermarking process traces back to music piracy source

A new watermarking process that leaves a breadcrumb trail leading to anyone that has illegally distributed an audio file, without affecting the quality of the original material, has been developed by researchers from Deakin University, Australia and Aizu University, Japan.

The method is also robust to de-synchronisation attacks common in other methods developed previously.

The new method is the most robust yet, with almost 100 per cent detection rate under the most common attacks, according to Lead researcher on the project, Deakin University School of Information Technology Associate Professor Yong Xiang.

“Pirating is a major issue for the music industry world-wide and in the global context, 95 per cent of music downloads are illegal,” Professor Xiang said.

Professor Xiang said watermarking worked by hiding watermark data, such as publisher name, signature, logo, ID number and other user information into the actual multimedia object without affecting its normal usage.

 “Watermarking technology can be used to prove copyright ownership, trace the source of illegal distribution and verify the authenticity of files,” he continued.

Digital watermarking technology can help tackle piracy by acting as a deterrent to malicious distribution of content by accurately identifying the source of leaked contents. This can provide the irrefutable evidence needed for legal action.

“What we did was to enable music file owners and relevant law enforcement authorities to use a secret key to extract the watermark data from the watermarked multimedia object,” Professor Xiang said.

 The technology also has other uses, such as authenticating content, filtering content such as blocking and triggering of actions, broadcast monitoring or locating content online.

The proposed method is robust to de-synchronisation attacks such as pitch-scaling, time-scaling, and jitter attacks, along with other conventional attacks such as re-quantisation, noise, amplitude, compression, re-sampling and filtering.

But of course, the technology could still be circumvented by recording a live playback as many pirates allegedly do.

Then there is the danger that the key itself is leaked on the internet enabling pirates to remove the watermark themselves.

The research is to be published in the latest edition of IEEE/ACM Transactions on Audio, Speech, and Language Processing.

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them