StirMark benchmark: audio watermarking attacks

We briefly present the architecture of a public automated evaluation service we are developing for still images, sound and video. We also detail new tests that will be included in this platform. The set of tests is related to audio data and addresses the usual equalisation and normalisation but also time stretching, pitch shifting and specially designed audio attack algorithms. These attacks are discussed and results on watermark attacks and perceived quality after applying the attacks are provided