Unified Speech and Noise Corpora for F0 Pitch Estimation Evaluation
July 5, 2026
This dataset consolidates CMU-ARCTIC, MOCHA-TIMIT, and PTDB-TUG with various noise corpora into standardized JBOF dataframes. It provides a unified framework for evaluating fundamental frequency estimation algorithms across multiple speech and environmental noise environments.
HOW THIS AFFECTS YOU
●
researcherYou can use these standardized dataframes to benchmark pitch tracking algorithms against diverse noise profiles.