Quality assessment in the streaming media industry has ma-tured to the stage that it encompasses not only traditional notions of pixel and audio sample integrity but also file for-mat 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 incorpo-rates the idea of learning features which are robust to coding artefacts to establish robust fingerprints for A/V Sync mea-surement. Results from large scale testing of 30,000 clips from YouTube show why measurement of A/V Sync is im-portant for file based video repositories and highlights issues that can now be addressed quantitatively. Index Terms — Lip sync, audio/video synchronisation, LDA, h.264, vp9

A novel framework for background music identification is proposed in this paper. Given a piece of audio signals that mixes background music with speech/noise, we identify the music part with source music data. Conventional methods that take the whole audio signals for identification are inap-propriate in terms of efficiency and accuracy. In our frame-work, the audio content is filtered through speech center cancellation and noise removal to extract clear music seg-ments. To identify these music segments, we use a compact feature representation and efficient similarity measurement based on the min-hash theory. The results of experiments on the RWC music database show a promising direction. Index Terms — fingerprint identification, copy detec-tion, content-based retrieval, min-hash