Traditional Machine Learning for Pitch Detection
This work addresses pitch detection for speech processing applications, but it is incremental as it applies traditional methods to a problem recently approached with deep learning.
The paper tackled pitch detection in speech processing by framing voicing detection as classification and F0 contour estimation as regression using traditional machine learning methods, achieving a 20% relative reduction in voicing errors with K-means and 45% with a Multi-Layer Perceptron, and showing perceptual preferences in synthesis tasks.
Pitch detection is a fundamental problem in speech processing as F0 is used in a large number of applications. Recent articles have proposed deep learning for robust pitch tracking. In this paper, we consider voicing detection as a classification problem and F0 contour estimation as a regression problem. For both tasks, acoustic features from multiple domains and traditional machine learning methods are used. The discrimination power of existing and proposed features is assessed through mutual information. Multiple supervised and unsupervised approaches are compared. A significant relative reduction of voicing errors over the best baseline is obtained: 20% with the best clustering method (K-means) and 45% with a Multi-Layer Perceptron. For F0 contour estimation, the benefits of regression techniques are limited though. We investigate whether those objective gains translate in a parametric synthesis task. Clear perceptual preferences are observed for the proposed approach over two widely-used baselines (RAPT and DIO).