Abstract | ||
---|---|---|
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. |
Year | Venue | Keywords |
---|---|---|
2015 | 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP) | Recurrent Neural Networks, Polyphonic Music Transcription, Music Language Models |
Field | DocType | Volume |
Transcription (linguistics),Pattern recognition,Computer science,Recurrent neural network,Speech recognition,Artificial intelligence,Generative grammar,Classifier (linguistics),Hidden Markov model,Artificial neural network,Language model | Conference | abs/1411.1623 |
ISSN | Citations | PageRank |
1520-6149 | 10 | 0.61 |
References | Authors | |
16 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Siddharth Sigtia | 1 | 127 | 8.56 |
Emmanouil Benetos | 2 | 557 | 52.48 |
Nicolas Boulanger-Lewandowski | 3 | 578 | 36.23 |
Tillman Weyde | 4 | 126 | 27.15 |
Artur S. D'avila Garcez | 5 | 431 | 63.57 |
Simon Dixon | 6 | 1164 | 107.57 |