Sigtia, S., Benetos, E., Boulanger-Lewandowski, N., Weyde, T., Garcez, A. & Dixon, S. (2015). A Hybrid Recurrent Neural Network For Music Transcription. Paper presented at the 40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015, 19-04-2015 - 24-04-2015, Brisbane, Australia.
- Accepted Version
Download (94kB) | Preview
We investigate the problem of incorporating higher-level symbolic score-like information into Automatic Music Transcription (AMT) systems to improve their performance. We use recurrent neural networks (RNNs) and their variants as music language models (MLMs) and present a generative architecture for combining these models with predictions from a frame level acoustic classifier. We also compare different neural network architectures for acoustic modeling. The proposed model computes a distribution over possible output sequences given the acoustic input signal and we present an algorithm for performing a global search for good candidate transcriptions. The performance of the proposed model is evaluated on piano music from the MAPS dataset and we observe that the proposed model consistently outperforms existing transcription methods.
|Item Type:||Conference or Workshop Item (Paper)|
|Additional Information:||© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.|
|Uncontrolled Keywords:||Recurrent Neural Networks, Polyphonic Music Transcription, Music Language Models|
|Subjects:||M Music and Books on Music
T Technology > TA Engineering (General). Civil engineering (General)
|Divisions:||School of Informatics|
Actions (login required)
Downloads per month over past year