Robust speech dereverberation based on non-negativity and sparse nature of speech spectrograms

By Hirokazu Kameoka, Tomohiro Nakatani, Takuya Yoshioka. 2009

87 citations recorded by the source corpus.

Subfields: Computational Mechanics, 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 Computational Mechanics

1. Nasser Mohammadiha
2. Rita Singh
3. Bhiksha Raj
4. Paris Smaragdis
5. Simon Doclo
6. Deliang Wang
7. Donald S. Williamson
8. Yi Luo

In Signal Processing

1. Nasser Mohammadiha
2. Paris Smaragdis
3. Rita Singh
4. Bhiksha Raj
5. Simon Doclo
6. Deliang Wang
7. Tuomas Virtanen
8. John R. Hershey

Citations in the corpus

• Speech Dereverberation Using Non-Negative Convolutive Transfer Function and Spectro-Temporal Modeling (2016, methodology): “1 shows an example to visualize the quality of the approximation for the N-CTF model when using h ( k, τ ) = | h c ( k, τ ) | to approximate the reverberant magnitude spectro-gram Y in (5), where a measured RIR ( T 60 ≈ 680 ms, direct-to-reverberation ratio (DRR) around 0 dB) is used.”
• Latent-variable decomposition based dereverberation of monaural and multi-channel signals (2010, methodology): “In [5], a non-negative matrix factorization based method is presented that only makes assumptions about the sparsity of the distribution of energy in spectral bands, and is similar in concept to the approach presented in this paper.”