@inproceedings{5163d9f5064d4c3aa289de5b40499ebc,
title = "Neural network alternatives toconvolutive audio models for source separation",
abstract = "Convolutive Non-Negative Matrix Factorization model factorizes a given audio spectrogram using frequency templates with a temporal dimension. In this paper, we present a convolutional auto-encoder model that acts as a neural network alternative to convolutive NMF. Using the modeling flexibility granted by neural networks, we also explore the idea of using a Recurrent Neural Network in the encoder. Experimental results on speech mixtures from TIMIT dataset indicate that the convolutive architecture provides a significant improvement in separation performance in terms of BSS eval metrics.",
keywords = "Auto-encoders, Convolutive models, Deep learning, Source separation",
author = "Shrikant Venkataramani and Cem Subakan and Paris Smaragdis",
note = "Funding Information: This work was supported by NSF grant 1453104. Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 ; Conference date: 25-09-2017 Through 28-09-2017",
year = "2017",
month = dec,
day = "5",
doi = "10.1109/MLSP.2017.8168108",
language = "English (US)",
series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",
publisher = "IEEE Computer Society",
pages = "1--6",
editor = "Naonori Ueda and Jen-Tzung Chien and Tomoko Matsui and Jan Larsen and Shinji Watanabe",
booktitle = "2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 - Proceedings",
}