Jazz Bass Transcription Using a U-Net Architecture

In this paper, we adapt a recently proposed U-net deep neural network architecture from melody to bass transcription. We investigate pitch shifting and random equalization as data augmentation techniques. In a parameter importance study, we study the influence of the skip connection strategy between the encoder and decoder layers, the data augmentation strategy, as well as of the overall model capacity on the system's performance. Using a training set that covers various music genres and a validation set that includes jazz ensemble recordings, we obtain the best transcription performance for a downscaled version of the reference algorithm combined with skip connections that transfer intermediate activations between the encoder and decoder. The U-net based method outperforms previous knowledge-driven and data-driven bass transcription algorithms by around five percentage points in overall accuracy. In addition to a pitch estimation improvement, the voicing estimation performance is clearly enhanced.

Automated summary

In this paper, the U-net deep neural network architecture is adapted for bass transcription, with pitch shifting and random equalization investigated as data augmentation techniques. The study examines the importance of skip connections between encoder and decoder layers, data augmentation strategies, and overall model capacity on system performance. Results show that using skip connections with training and validation sets can improve bass transcription performance.