On Local Keys, Modulations, and Tonicizations
Bibliographic record
Abstract
Throughout the common-practice period (1650-1900), it is customary to find changes of musical key within a piece of music. In current music theory terminology, the concepts of modulation and tonicization are helpful to explain many of these changes of key. Conversely, in computational musicology and music information retrieval, the preferred way to denote changes of key are local key features, which are oftentimes predicted by computational models. Therefore, the three concepts, local keys, modulations, and toniciza-tions describe changes of key. What is, however, the relationship between the local keys, modulations, and tonicizations of the same musical fragment? In this paper, we contribute to this research question by 1) reviewing the current methods of local-key estimation, 2) providing a new dataset with annotated modulations and tonicizations, and 3) applying all the annotations (i.e., local keys, modulations, and tonicizations) in an experiment that connects the three concepts together. In our experiment, instead of assuming the music-theoretical meaning of the local keys predicted by an algorithm, we evaluate whether these coincide better with the modulation or tonicization annotations of the same musical fragment. Three existing models of symbolic local-key estimation, together with the annotated modulations and tonicizations of five music theory textbooks are considered during our evaluation. We provide the methodology of our experiment and our dataset (available at https://github.com/DDMAL/key_modulation_dataset) to motivate future research in the relationship between local keys, modulations, and tonicizations. * Both authors contributed equally to this work. • Applied computing → Sound and music computing; Fine arts; • Information systems → Retrieval tasks and goals; Evaluation of retrieval results.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".