An End-to-End Neural Network for Polyphonic Music Transcription
We present a neural network model for polyphonic music transcription. The architecture of the proposed model is analogous to speech recognition systems and comprises an acoustic model and a music language mode}. The acoustic model is a neural network used for estimating the probabilities of pitches in a frame of audio. The language model is a recurrent neural network that models the correlations between pitch combinations over time. The proposed model is general and can be used to transcribe polyphonic music without imposing any constraints on the polyphony or the number or type of instruments. The acoustic and language model predictions are combined using a probabilistic graphical model. Inference over the output variables is performed using the beam search algorithm. We investigate various neural network architectures for the acoustic models and compare their performance to two popular state-of-the-art acoustic models. We also present an efficient variant of beam search that improves performance and reduces run-times by an order of magnitude, making the model suitable for real-time applications. We evaluate the model's performance on the MAPS dataset and show that the proposed model outperforms state-of-the-art transcription systems.