Summary of Unsupervised Learning Of Harmonic Analysis Based on Neural Hsmm with Code Quality Templates, by Yui Uehara
Unsupervised Learning of Harmonic Analysis Based on Neural HSMM with Code Quality Templates
by Yui Uehara
First submitted to arxiv on: 7 Mar 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a method for unsupervised harmonic analysis using a hidden semi-Markov model (HSMM). The authors introduce chord quality templates that specify the probability of pitch class emissions given a root note and chord quality. Other probability distributions are automatically learned through unsupervised learning, addressing a challenge in existing research. Evaluation results show that while this method doesn’t yet match performance of supervised models with complex rule design, it has the advantage of not requiring labeled data or elaborate rules. Additionally, the authors demonstrate how to recognize tonics without prior knowledge based on Markov model transition probabilities. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about using a special type of computer program (called an HSMM) to analyze music patterns without needing any training data. The program uses “templates” that define what sounds are likely to come next in a song, given the starting note and chord quality. This method learns patterns on its own, which is helpful because it doesn’t require expensive training data or complicated rules. The authors also show how this method can identify the main theme of a piece without knowing anything about it beforehand. |
Keywords
» Artificial intelligence » Markov model » Probability » Supervised » Unsupervised
