Improved Method of Blind Speech Separation with Low Computational Complexity | Zendy

Kazunobu Kondo | Zendy; Yu Takahashi | Zendy; Seiichi Hashimoto | Zendy; Hiroshi Saruwatari | Zendy; Takanori Nishino | Zendy; Kazuya Takeda | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

Improved Method of Blind Speech Separation with Low Computational Complexity

Author(s) -

Kazunobu Kondo,

Yu Takahashi,

Seiichi Hashimoto,

Hiroshi Saruwatari,

Takanori Nishino,

Kazuya Takeda

Publication year - 2011

Publication title -

advances in acoustics and vibration

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 0.237

H-Index - 14

eISSN - 1687-627X

pISSN - 1687-6261

DOI - 10.1155/2011/765429

Subject(s) - computational complexity theory , anechoic chamber , computer science , distortion (music) , blind signal separation , cepstrum , frequency domain , speech recognition , selection (genetic algorithm) , speech enhancement , transfer function , signal to noise ratio (imaging) , algorithm , signal (programming language) , noise (video) , artificial intelligence , noise reduction , engineering , channel (broadcasting) , telecommunications , computer vision , amplifier , image (mathematics) , electrical engineering , bandwidth (computing) , programming language

A blind speech separation method with low computational complexity is proposed. This method consists of a combination of independent component analysis with frequency band selection, anda frame-wise spectral soft mask method based on an interchannel power ratio of tentative separated signals in the frequency domain. The soft mask cancels the transfer function between sources and separated signals. A theoretical analysis of selection criteria and the soft mask is given. Performance and effectiveness are evaluated via source separation simulations and a computational estimate, and experimental results show the significantly improved performance of the proposed method. The segmental signal-to-noise ratio achieves 7 [dB] and 3 [dB], and the cepstral distortion achieves 1 [dB] and 2.5 [dB], in anechoic and reverberant conditions, respectively. Moreover, computational complexity is reduced by more than 80% compared with unmodified FDICA

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research