SPICE: Self-supervised Pitch Estimation

TL;DR

This paper introduces SPICE, a self-supervised model for pitch estimation in monophonic audio that achieves accuracy comparable to supervised methods without needing labeled data.

Contribution

The paper presents a novel self-supervised approach for pitch estimation using constant-Q transform and a confidence head, eliminating the need for annotated datasets.

Findings

Achieves comparable accuracy to supervised models on clean and noisy audio.

Does not require large labeled datasets for training.

Effective in both voiced and unvoiced audio segments.

Abstract

We propose a model to estimate the fundamental frequency in monophonic audio, often referred to as pitch estimation. We acknowledge the fact that obtaining ground truth annotations at the required temporal and frequency resolution is a particularly daunting task. Therefore, we propose to adopt a self-supervised learning technique, which is able to estimate pitch without any form of supervision. The key observation is that pitch shift maps to a simple translation when the audio signal is analysed through the lens of the constant-Q transform (CQT). We design a self-supervised task by feeding two shifted slices of the CQT to the same convolutional encoder, and require that the difference in the outputs is proportional to the corresponding difference in pitch. In addition, we introduce a small model head on top of the encoder, which is able to determine the confidence of the pitch estimate, so as to distinguish between voiced and unvoiced audio. Our results show that the proposed method is able to estimate pitch at a level of accuracy comparable to fully supervised models, both on clean and noisy audio samples, although it does not require access to large labeled datasets.