Phase Processing for Single-Channel Speech Enhancement: History and recent advances

By Timo Gerkmann, Martin Krawczyk-Becker, Jonathan Le Roux. 2015

236 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Signal Processing

Citations in the corpus

Speech Denoising with Deep Feature Losses (2018, background): “This issue can be alleviated somewhat, but with increased computational cost and system complexity [12, 13, 14, 15, 16, 17, 18].”
Time-Frequency Masking in the Complex Domain for Speech Dereverberation and Denoising (2017, neutral): “More details about different phase enhancement techniques can be found in [11].”
End-to-End Speech Separation with Unfolded Iterative Phase Reconstruction (2018, methodology): “To improve the consistency, one stream of research is focused on iterative methods such as the classic Griffin-Lim algorithm [14], multiple input spectrogram inverse (MISI) [15], ISSIR [16], and consistent Wiener filtering [17], which can recover the clean phase to some extent starting from the mixt”
Supervised Speech Separation Based on Deep Learning: An Overview (2017, unknown): “First, temporal mapping is a welcome addition to the list of supervised separation approaches and provides a unique perspective to phase enhancement [50],”
Speech dereverberation and denoising using complex ratio masks (2017, background): “Different phase enhancement approaches are discussed in [10, 11, 12].”
The Phasebook: Building Complex Masks via Discrete Representations for Source Separation (2019, background): “Because the noisy phases are typically inconsistent with the estimated magnitudes [12, 13], the reconstructed timedomain signal has a different magnitude spectrogram from the estimated one.”
End-to-End Waveform Utterance Enhancement for Direct Evaluation Metrics Optimization by Fully Convolutional Neural Networks (2017, background): “However, the enhanced magnitude spectrogram 𝐸 may not correspond to the STFT of any time-domain signal anymore [40, 55].”
Alternative Objective Functions for Deep Clustering (2018, methodology): “Conventionally, iterative methods, such as the Grifﬁn-Lim algorithm [19], can recover the clean phase to some extent starting from the noisy phase and a good estimated magnitude by iteratively performing STFT and iSTFT [20].”
Deep Learning Based Phase Reconstruction for Speaker Separation: A Trigonometric Perspective (2018, background): “3), and is mathematically related to, log magnitude [11], [19]–[21].”