On the optimality of ideal binary time-frequency masks

By Yipeng Li, Deliang Wang. 2008

171 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, 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 Cognitive Neuroscience

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”
On Training Targets for Supervised Speech Separation (2014, background): “For example, dB maximizes the SNR gain of the IBM [20].”
An algorithm to improve speech recognition in noise for hearing-impaired listeners. (2013, methodology): “The term also reflects the fact that the IBM provides the optimal SNR gain of all binary T-F masks under certain conditions (Li and Wang, 2009).”
Speaker-Independent Speech Separation With Deep Attractor Network (2017, background): “2R1 FT , and calculated by ^s i = xm i (3) subject to XC i=1 m i = 1 (4) where is element-wise multiplication and 1 2R1 FT denotes all-one vector. Commonly used masks include ideal binary mask (IBM) [38], ideal ratio mask (IRM) [39], and ‘wiener-ﬁlter’ like mask (WFM) [40]: 8 >> >> >&”
Speech intelligibility in reverberation with ideal binary masking: effects of early reflections and signal-to-noise ratio threshold. (2013, background): “…0 dB can lead to improved intelligibility by better restoring the spectro-temporal modulation pattern of target speech (Brungart et al. , 2006; Li and Loizou, 2008; Wang et al. , 2009; Kjems et al. , 2009); in spite of the fact that such masks are suboptimal in terms of SNR (Li and Wang, 2009).”
Phase-sensitive and recognition-boosted speech separation using deep recurrent neural networks (2015, background): “Various “ideal” masks have been used for a∗; the most common have been the so-called “ideal binary mask”(IBM) [17], and the “ideal ratio mask”(IRM) [18].”