AI TrackerFrontier desk · AEST
Back to news
normalaitech

Revisiting Input Time-frequency Representations in Multi-pitch Estimation for Vocal Ensembles

arXiv cs.AI · 3 Oct 2026, 03:35 AEST

Multi-pitch estimation in vocal ensembles faces challenges due to overlapping pitch ranges and closely spaced fundamental frequencies. A study finds that linear short-time Fourier transform (STFT) outperforms harmonic constant-Q transform (HCQT) while reducing feature-extraction costs.

Key facts

Read aloud

Multi-pitch estimation in vocal ensembles faces challenges due to overlapping pitch ranges and closely spaced fundamental frequencies. A study finds that linear short-time Fourier transform STFT outperforms harmonic constant-Q transform HCQT while reducing feature-extraction costs. Multi-pitch estimation in vocal ensembles is challenging because singers occupy overlapping pitch ranges and often sing at closely spaced fundamental frequencies, causing their harmonics to overlap in time-frequency representations. Existing models commonly use harmonic constant-Q transform HCQT -based representations to provide frequency-adaptive resolution, at the cost of expensive feature extraction when training mixtures are generated on the fly. Despite its fixed frequency resolution and the absence of a pitch-aligned input grid, the linear STFT outperforms HCQT while substantially reducing feature-extraction cost. Further analysis shows that a longer analysis window or broader spectral coverage provides no additional improvement, while restricting the input to the predicted pitch range reduces the advantage of the linear STFT.

Sources in this cluster