Multi-label Ferns for Efficient Recognition of Musical Instruments in Recordings
This work addresses efficient recognition of musical instruments in recordings, which is an incremental improvement for audio analysis applications.
The paper tackled the problem of automatic classification of musical instruments in audio recordings by introducing multi-label ferns, resulting in much faster classification, higher F-score, and substantial reduction in model size compared to binary random ferns.
In this paper we introduce multi-label ferns, and apply this technique for automatic classification of musical instruments in audio recordings. We compare the performance of our proposed method to a set of binary random ferns, using jazz recordings as input data. Our main result is obtaining much faster classification and higher F-score. We also achieve substantial reduction of the model size.