# Lightweight Self-Supervised Detection of Fundamental Frequency and Accurate Probability of Voicing in Monophonic Music

arXiv: https://arxiv.org/abs/2601.11768 · HTML: https://mlsanity.com/paper/2601.11768/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 22 references that are in the map.
- First posted on arXiv 16 Jan 2026.

## Builds on (most cited first, 9 of 9)

- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [DDSP: Differentiable Digital Signal Processing](https://mlsanity.com/paper/2001.04643/) (2020)
- [CREPE: A Convolutional Representation for Pitch Estimation](https://mlsanity.com/paper/1802.06182/) (2018)
- [SPICE: Self-supervised Pitch Estimation](https://mlsanity.com/paper/1910.11664/) (2019)
- [A Review of Differentiable Digital Signal Processing for Music & Speech Synthesis](https://mlsanity.com/paper/2308.15422/) (2023)
- [PESTO: Pitch Estimation with Self-supervised Transposition-equivariant Objective](https://mlsanity.com/paper/2309.02265/) (2023)
- [Cross-domain Neural Pitch and Periodicity Estimation](https://mlsanity.com/paper/2301.12258/) (2023)
- [DEEPF0: End-To-End Fundamental Frequency Estimation for Music and Speech Signals](https://mlsanity.com/paper/2102.06306/) (2021)
- [PESTO: Real-Time Pitch Estimation with Self-supervised Transposition-equivariant Objective](https://mlsanity.com/paper/2508.01488/) (2025)

## Abstract (arXiv)

Reliable fundamental frequency (F 0) and voicing estimation is essential for neural synthesis, yet many pitch extractors depend on large labeled corpora and degrade under realistic recording artifacts. We propose a lightweight, fully self-supervised framework for joint F 0 estimation and voicing inference, designed for rapid single-instrument training from limited audio. Using transposition-equivariant learning on CQT features, we introduce an EM-style iterative reweighting scheme that uses Shift Cross-Entropy (SCE) consistency as a reliability signal to suppress uninformative noisy/unvoiced frames. The resulting weights provide confidence scores that enable pseudo-labeling for a separate lightweight voicing classifier without manual annotations. Trained on MedleyDB and evaluated on MDB-stem-synth ground truth, our method achieves competitive cross-corpus performance (RPA 95.84, RCA 96.24) and demonstrates cross-instrument generalization.