Different Pitch Distribution

Research on pitch distribution focuses on understanding and modeling how pitch varies across musical performances and different vocal types. Current efforts involve developing algorithms, such as dynamic programming and neural networks (including architectures like Harmonic Dilated Convolution and Recurrent Neural Networks), to accurately transcribe and analyze pitch data, addressing challenges like gender bias in transcription systems and modeling harmonic structures. This work is significant for improving music transcription accuracy, enhancing music information retrieval systems, and providing insights into musical performance and vocal production.