Consistent independent low-rank matrix analysis for determined blind source separation

Abstract

Independent low-rank matrix analysis (ILRMA) is the state-of-the-art algorithm for blind source separation (BSS) in the determined situation (the number of microphones is greater than or equal to that of source signals). ILRMA achieves a great separation performance by modeling the power spectrograms of the source signals via the nonnegative matrix factorization (NMF). Such a highly developed source model can solve the permutation problem of the frequency-domain BSS to a large extent, which is the reason for the excellence of ILRMA. In this paper, we further improve the separation performance of ILRMA by additionally considering the general structure of spectrograms, which is called consistency, and hence, we call the proposed method Consistent ILRMA. Since a spectrogram is calculated by an overlapping window (and a window function induces spectral smearing called main- and side-lobes), the time-frequency bins depend on each other. In other words, the time-frequency components are related to each other via the uncertainty principle. Such co-occurrence among the spectral components can function as an assistant for solving the permutation problem, which has been demonstrated by a recent study. On the basis of these facts, we propose an algorithm for realizing Consistent ILRMA by slightly modifying the original algorithm. Its performance was extensively evaluated through experiments performed with various window lengths and shift lengths. The results indicated several tendencies of the original and proposed ILRMA that include some topics not fully discussed in the literature. For example, the proposed Consistent ILRMA tends to outperform the original ILRMA when the window length is sufficiently long compared to the reverberation time of the mixing system.

InfoMasker: Preventing Eavesdropping Using Phoneme-Based Noise

“…In the case study, the recordings from the Huawei Mate30Pro have two channels, so we test the effectiveness of BSS on these recordings. We test five BSS algorithms: AuxIVA [55], ConsistentILRMA [56], FastMNMF [57], LaplaceFDICA [58], and t-ILRMA [59]. The results are shown in Table 28(bottom).…”

Advances in Time-Frequency Analysis for Blind Source Separation: Challenges, Contributions, and Emerging Trends

“…After extracting features to separate the source signals, we utilize methods such as STFT, CWT, WVD, and FrFT for separating and recovering the source signals from both the original mixed signals and the mixed signals with added noise [38], [39].This allows us to compare the significant role of time-frequency analysis in the process of source signal recovery. From Figures 19-22, it can be observed that STFT WVD exhibit relatively better recovery results, whereas CWT shows inferior performance primarily due to the impact of cross-terms, leading to significant deviations.…”

Rolling Bearing Composite Fault Diagnosis Method Based on Enhanced Harmonic Vector Analysis

“…By considering the low-rank time-frequency structure of the source signal, IVA is further extended to an independent low-rank matrix analysis (ILRMA). ILRMA uses non-negative matrix factorization (NMF) to model the power spectrum of the source signals and achieves the best separation performance [21]. On the basis of ILRMA, Kohei Yatabe et al proposed harmonic vector analysis (HVA) to solve the problem of BSS of convoluted mixed speech signals [22].…”