A Hybrid Recurrent Neural Network For Music Transcription
This work addresses the challenge of accurate music transcription for applications in music analysis and generation, representing an incremental improvement over prior methods.
The paper tackles the problem of improving Automatic Music Transcription (AMT) by incorporating higher-level symbolic score-like information, resulting in a model that consistently outperforms existing methods on piano music from the MAPS dataset.
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.