Assessment of Audio/Video synchronisation in streaming media

Quality assessment in the streaming media industry has matured to the stage that it encompasses not only traditional notions of pixel and audio sample integrity but also file format consistency and the media consumption experience itself. Audio/Video synchronisation has already been established as an associated measure of media quality and is well known in video conferencing and movie streaming applications. This paper presents a new system for the assessment of audio and video synchronisation in a media file. The system incorporates the idea of learning features which are robust to coding artefacts to establish robust fingerprints for A/V Sync measurement. Results from large scale testing of 30,000 clips from YouTube show why measurement of A/V Sync is important for file based video repositories and highlights issues that can now be addressed quantitatively.