convolutive speech blind source separation by time –frequency masking
عنوان مقاله: convolutive speech blind source separation by time –frequency masking
شناسه ملی مقاله: ICEEE05_434
منتشر شده در پنجمین کنفرانس ملی مهندسی برق و الکترونیک ایران در سال 1392
شناسه ملی مقاله: ICEEE05_434
منتشر شده در پنجمین کنفرانس ملی مهندسی برق و الکترونیک ایران در سال 1392
مشخصات نویسندگان مقاله:
Atefe Soltani - Najafabad Branch,Islamic Azad University
Hamid Mahmoodian - Najafabad Branch,Islamic Azad University
خلاصه مقاله:
Atefe Soltani - Najafabad Branch,Islamic Azad University
Hamid Mahmoodian - Najafabad Branch,Islamic Azad University
This paper presents a T-F masking method for convolutive blind source separation based on hermitian angle concept. The hermitian angle is calculated between T-Fdomain mixture vector and reference vector. Two different reference vectors are assumed for calculating two different hermitian angles, and then these angles are clustered with kmeans or FCM method to estimate unmixing masks. The well known permutation problem is solved based on k-means clustering of estimated masks which are partitioned to small groups. The experiment results show that separation performance for two different reference vectors is better than that for only one reference vector
کلمات کلیدی: blind source separation (BSS); sparsity ; w-disjoint orthogonality ; hermitian angle;time-frequency masking
صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/219750/