On Ideal Binary Mask As the Computational Goal of Auditory Scene Analysis

By Deliang Wang. 2005

658 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Signal Processing

Citations in the corpus

Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation (2018, background): “Moreover, the separation accuracy of Conv-TasNet surpasses the performance of ideal time-frequency magnitude masks, including the ideal binary mask (IBM [34]), ideal ratio mask (IRM [35], [36]), and Winener filter-like mask (WFM [37]) in both signal-to-distortion ratio (SDR) and subjective (mean”
A Spectral Masking Approach to Noise-Robust Speech Recognition Using Deep Neural Networks (2014, positive): “With stereo data, Ideal Binary Masks (IBMs) [17] have been shown to substantially improve the intelligibility of speech with background noise [18].”
Alzheimer's Dementia Recognition through Spontaneous Speech: The ADReSS Challenge (2020, neutral): “MRCG features are based on cochleagrams [19].”
Single-Channel Multi-Speaker Separation Using Deep Clustering (2016, background): “Silence is deﬁned using a threshold on the energy relative to the maximum of the mixture.”
Exploring Monaural Features for Classification-Based Speech Segregation (2013, background): “Computational auditory scene analysis (CASA), which is inspired by Bregman’s account of auditory scene analysis (ASA) [2], has shown considerable promise in the last decade.”
Deep Learning for Environmentally Robust Speech Recognition (2017, background): “1) Masks: Two most commonly used masks in the literature include: binary-based mask [85] and ratio-based mask [86].”
A Deep Ensemble Learning Method for Monaural Speech Separation (2016, background): “For the IBM [29], a T-F unit is assigned 1, if the signal-to-noise ratio (SNR) within the unit exceeds a local criterion, indicating target dominance.”
Speech intelligibility in reverberation with ideal binary masking: effects of early reflections and signal-to-noise ratio threshold. (2013, background): “The ideal time-frequency binary mask (IBM) has been proposed as a performance upper bound for the separation capacity of CASA algorithms (Wang, 2005).”
NADiffuSE: Noise-aware Diffusion-based Model for Speech Enhancement (2023, methodology): “The discriminative methods learn non-linear mappings[4] or estimate time-frequency masks [5]–[7] through annotated speech data pairs, but suffer from non-linear artifacts and poor generalizations [8].”
On the Improvement of Singing Voice Separation for Monaural Recordings Using the MIR-1K Dataset (2010, methodology): “optimizing this decision process to create an ideal binary mask [11].”
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement (2020, background): “In the masking family, ideal binary mask (IBM) [20], ideal ratio mask (IRM) [10] and spectral magnitude mask (SMM) [21] only use the magnitude between clean speech and mixture speech, ignoring the phase information.”
Fullsubnet: A Full-Band and Sub-Band Fusion Model for Real-Time Single-Channel Speech Enhancement (2020, background): “Learning target is the spectral feature of clean speech or a certain mask (e.g., Ideal Binary Mask [8], Ideal Ratio Mask [9], complex Ideal Ratio Mask (cIRM) [10]).”
Reasons why current speech-enhancement algorithms do not improve speech intelligibility and suggested solutions (2011, background): “In this Appendix, we provide analytical proof that the IdBM criterion is optimal in that it maximizes the geometric average of the spectral SNRs.”
A Consolidated Perspective on Multimicrophone Speech Enhancement and Source Separation (2017, neutral): “In [265], a soft mask is proposed based on a mapping between the ITD values and the relative contribution of the target source.”
An algorithm to improve speech recognition in noise for hearing-impaired listeners. (2013, methodology): “Another limitation is that of configuration stationarity: All spatial filtering methods operate on the premise of a fixed-source configuration (Wang, 2005).”