Paper ID | AUD-19.1 | ||
Paper Title | PHASE RECOVERY WITH BREGMAN DIVERGENCES FOR AUDIO SOURCE SEPARATION | ||
Authors | Paul Magron, Pierre-Hugo Vial, IRIT, Université de Toulouse, CNRS, France; Thomas Oberlin, ISAE-SUPAERO, Université de Toulouse, France; Cédric Févotte, IRIT, Université de Toulouse, CNRS, France | ||
Session | AUD-19: Audio and Speech Source Separation 6: Topics in Source Separation | ||
Location | Gather.Town | ||
Session Time: | Thursday, 10 June, 13:00 - 13:45 | ||
Presentation Time: | Thursday, 10 June, 13:00 - 13:45 | ||
Presentation | Poster | ||
Topic | Audio and Acoustic Signal Processing: [AUD-SEP] Audio and Speech Source Separation | ||
IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
Abstract | Time-frequency audio source separation is usually achieved by estimating the short-time Fourier transform (STFT) magnitude of each source, and then applying a phase recovery algorithm to retrieve time-domain signals. In particular, the multiple input spectrogram inversion (MISI) algorithm has shown good performance in several recent works. This algorithm minimizes a quadratic reconstruction error between magnitude spectrograms. However, this loss does not properly account for some perceptual properties of audio, and alternative discrepancy measures such as beta-divergences have been preferred in many settings. In this paper, we propose to reformulate phase recovery in audio source separation as a minimization problem involving Bregman divergences. To optimize the resulting objective, we derive a projected gradient descent algorithm. Experiments conducted on a speech enhancement task show that this approach outperforms MISI for several alternative losses, which highlights their relevance for audio source separation applications. |