Science – Future of Lithuania / Mokslas – Lietuvos Ateitis, Vol 3, No 1 (2011)

Speaker Localization and Speech Separationin Two Echoic Mixtures

Włodzimierz Kasprzak
Ning Ding
Nozomu Hamada

Abstract


We are developing two crucial improvements on the time-frequency masking approach to the blind speech separation of underdetermined mixtures when processing anechoic and echoic mixtures. First, the proposed method copes with the usually large amount of delay estimation error that appears in a low frequency band. This step generates a restrictive mask for phase delays on the basis of local and global energy distribution analysis. This mask allows the selected cells to contribute to the orientation histogram. Second, the strong WDO assumption (disjoint orthogonal frequency domain) is relaxed by allowing some frequency bins to be shared by both sources. By detecting fundamental frequencies of speakers at instantaneous time points, mask creation is supported by exploring their harmonic frequencies. The proposed method is proved to be effective and reliable in conducting experiments with both simulated and real-life mixtures.

Article in English


Article in: English

Article published: 2011-08-22

Keyword(s): blind source separation; histogram clustering; spectrogram analysis; speech reconstruction; time-frequency masking

DOI: 10.3846/mla.2011.009

Full Text: PDF pdf

Science – Future of Lithuania / Mokslas – Lietuvos Ateitis ISSN 2029-2341, eISSN 2029-2252
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 License.