Publication Details
Compact Network for Speakerbeam Target Speaker Extraction
Žmolíková Kateřina, Ing., Ph.D. (DCGM FIT BUT)
Ochiai Tsubasa (NTT)
Kinoshita Keisuke (NTT)
Araki Shoko (NTT)
Nakatani Tomohiro (NTT)
Target speech extraction, Neural network, Adaptation, Auxiliary feature, Speech enhancement
Speech separation that separates a mixture of speech signals into each of its sources has been an active research topic for a long time and has seen recent progress with the advent of deep learning. A related problem is target speaker extraction, i.e. extraction of only speech of a target speaker out of a mixture, given characteristics of his/her voice. We have recently proposed SpeakerBeam, which is a neural network-based target speaker extraction method. Speaker- Beam uses a speech extraction network that is adapted to the target speaker using auxiliary features derived from an adaptation utterance of that speaker. Initially, we implemented SpeakerBeam with a factorized adaptation layer, which consists of several parallel linear transformations weighted by weights derived from the auxiliary features. The factorized layer is effective for target speech extraction, but it requires a large number of parameters. In this paper, we propose to simply scale the activations of a hidden layer of the speech extraction network with weights derived from the auxiliary features. This simpler approach greatly reduces the number of model parameters by up to 60%, making it much more practical, while maintaining a similar level of performance. We tested our approach on simulated and real noisy and reverberant mixtures, showing the potential of SpeakerBeam for real-life applications. Moreover, we showed that speech extraction performance of SpeakerBeam compares favorably with that of a state-of-the-art speech separation method with a similar network configuration.
@INPROCEEDINGS{FITPUB12096,
author = "Marc Delcroix and Kate\v{r}ina \v{Z}mol\'{i}kov\'{a} and Tsubasa Ochiai and Keisuke Kinoshita and Shoko Araki and Tomohiro Nakatani",
title = "Compact Network for Speakerbeam Target Speaker Extraction",
pages = "6965--6969",
booktitle = "Proceedings of ICASSP",
year = 2019,
location = "Brighton, GB",
publisher = "IEEE Signal Processing Society",
ISBN = "978-1-5386-4658-8",
doi = "10.1109/ICASSP.2019.8683087",
language = "english",
url = "https://www.fit.vut.cz/research/publication/12096"
}