In this work, we propose a novel consistency-preserving loss function for recovering the phase information in the context of phase reconstruction (PR) and speech enhancement (SE). Different from conventional techniques that directly estimate the phase using a deep model, our idea is to exploit ad-hoc constraints to directly generate a consistent pair of magnitude and phase. Specifically, the proposed loss forces a set of complex numbers to be a consistent short-time Fourier transform (STFT) representation, i.e., to be the spectrogram of a real signal. Our approach thus avoids the difficulty of estimating the original phase, which is highly unstructured and sensitive to time shift. The influence of our proposed loss is first assessed on a PR task, experimentally demonstrating that our approach is viable. Next, we show its effectiveness on an SE task, using both the VB-DMD and WSJ0-CHiME3 data sets. On VB-DMD, our approach is competitive with conventional solutions. On the challenging WSJ0-CHiME3 set, the proposed framework compares favourably over those techniques that explicitly estimate the phase.

Ku P.-J., Ho C.-W., Yen H., Siniscalchi S.M., Lee C.-H. (2025). An Explicit Consistency-Preserving Loss Function for Phase Reconstruction and Speech Enhancement. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 1-5). Institute of Electrical and Electronics Engineers Inc. [10.1109/ICASSP49660.2025.10887812].

An Explicit Consistency-Preserving Loss Function for Phase Reconstruction and Speech Enhancement

Siniscalchi S. M.
Supervision
;
2025-01-01

Abstract

In this work, we propose a novel consistency-preserving loss function for recovering the phase information in the context of phase reconstruction (PR) and speech enhancement (SE). Different from conventional techniques that directly estimate the phase using a deep model, our idea is to exploit ad-hoc constraints to directly generate a consistent pair of magnitude and phase. Specifically, the proposed loss forces a set of complex numbers to be a consistent short-time Fourier transform (STFT) representation, i.e., to be the spectrogram of a real signal. Our approach thus avoids the difficulty of estimating the original phase, which is highly unstructured and sensitive to time shift. The influence of our proposed loss is first assessed on a PR task, experimentally demonstrating that our approach is viable. Next, we show its effectiveness on an SE task, using both the VB-DMD and WSJ0-CHiME3 data sets. On VB-DMD, our approach is competitive with conventional solutions. On the challenging WSJ0-CHiME3 set, the proposed framework compares favourably over those techniques that explicitly estimate the phase.
2025
979-8-3503-6874-1
Ku P.-J., Ho C.-W., Yen H., Siniscalchi S.M., Lee C.-H. (2025). An Explicit Consistency-Preserving Loss Function for Phase Reconstruction and Speech Enhancement. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 1-5). Institute of Electrical and Electronics Engineers Inc. [10.1109/ICASSP49660.2025.10887812].
File in questo prodotto:
File Dimensione Formato  
An_Explicit_Consistency-Preserving_Loss_Function_for_Phase_Reconstruction_and_Speech_Enhancement.pdf

Solo gestori archvio

Descrizione: main document
Tipologia: Versione Editoriale
Dimensione 391.13 kB
Formato Adobe PDF
391.13 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/679551
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact