An LSTM-Based Chord Generation System Using Chroma Histogram Representations
This is an incremental improvement for music generation tasks, addressing chord generation for monophonic melodies.
The paper tackled chord generation for monophonic symbolic melodies by using an LSTM-based model with chroma histogram representations, resulting in a system suitable for limited real-time use with diatonic generation and cadential chord relationships, though it did not meet state-of-the-art for long-term coherence.
This paper proposes a system for chord generation to monophonic symbolic melodies using an LSTM-based model trained on chroma histogram representations of chords. Chroma representations promise more harmonically rich generation than chord label-based approaches, whilst maintaining a small number of dimensions in the dataset. This system is shown to be suitable for limited real-time use. While it does not meet the state-of-the-art for coherent long-term generation, it does show diatonic generation with cadential chord relationships. The need for further study into chroma histograms as an extracted feature in chord generation tasks is highlighted.