Nonnegative tensor factorization with frequency modulation cues for\n blind audio source separation
Bibliographic record
Abstract
We present Vibrato Nonnegative Tensor Factorization, an algorithm for\nsingle-channel unsupervised audio source separation with an application to\nseparating instrumental or vocal sources with nonstationary pitch from music\nrecordings. Our approach extends Nonnegative Matrix Factorization for audio\nmodeling by including local estimates of frequency modulation as cues in the\nseparation. This permits the modeling and unsupervised separation of vibrato or\nglissando musical sources, which is not possible with the basic matrix\nfactorization formulation.\n The algorithm factorizes a sparse nonnegative tensor comprising the audio\nspectrogram and local frequency-slope-to-frequency ratios, which are estimated\nat each time-frequency bin using the Distributed Derivative Method. The use of\nlocal frequency modulations as separation cues is motivated by the principle of\ncommon fate partial grouping from Auditory Scene Analysis, which hypothesizes\nthat each latent source in a mixture is characterized perceptually by coherent\nfrequency and amplitude modulations shared by its component partials. We derive\nmultiplicative factor updates by Minorization-Maximization, which guarantees\nconvergence to a local optimum by iteration. We then compare our method to the\nbaseline on two separation tasks: one considers synthetic vibrato notes, while\nthe other considers vibrato string instrument recordings.\n
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
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 source (direct Gemma or distilled Codex), 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".