We have collected a database of musical features from radio broadcasts and CD collections (N> 10 5). The database poses a number of hard modelling challenges including: Segmentation problems and missing and wrong meta-data. We describe our efforts towards cleaning the data using probability density estimation. We train conditional densities for checking the relation between meta-data and music features, and un-conditional densities for spotting unlikely music features. We show that the rejected samples indeed represent various types of problems in the music data. The models may in some cases assist reconstruction of meta-data. 1.

Real-time classification applied to a vocal percussion signal holds potential as an interface for live musical control. In this article we propose a novel approach to resolving the tension between the needs for low-latency reaction and reliable classification, by deferring the final classification decision until after a response has been initiated. We introduce a new dataset of annotated human beatbox recordings, and use it to study the optimal delay for classification accuracy. We then investigate the effect of such delayed decision-making on the quality of the audio output of a typical reactive system, via a MUSHRA-type listening test. Our results show that the effect depends on the output audio type: for popular dance/pop drum sounds the acceptable delay is on the order of 12–35 ms. 1

The enormous growth of digital music databases has led to a comparable growth in the need for methods that help users organize and access such information. One area in particular that has seen much recent research activity is the use of automated techniques to describe audio content and to allow for its identification, browsing and retrieval. Conventional approaches to music content description rely on features characterizing the shape of the signal spectrum in relatively short-term frames. In the context of Automatic Speech Recognition (ASR), Hermansky [7] described an interesting alternative to short-term spectrum features, the TRAP-TANDEM approach which uses long-term band-limited features trained in a supervised fashion. We adapt this idea to the specific case of music signals and propose a generic system for the description of temporal patterns. The same system with different settings is able to extract features describing either timbre or rhythmic content. The quality of the generated features is demonstrated in a set of music retrieval experiments and compared to other state-of-the-art models. 1