How to train your speaker embeddings extractor

Mitchell Mclaren, Diego Castán, Mahesh Kumar Nandwana, Luciana Ferrer, Emre Yilmaz

With the recent introduction of speaker embeddings for text-independent speaker recognition, many fundamental questions require addressing in order to fast-track the development of this new era of technology. Of particular interest is the ability of the speaker embeddings network to leverage artificially degraded data at a far greater rate beyond prior technologies, even in the evaluation of naturally degraded data. In this study, we aim to explore some of the fundamental requirements for building a good speaker embeddings extractor. We analyze the impact of voice activity detection, types of degradation, the amount of degraded data, and number of speakers required for a good network. These aspects are analyzed over a large set of 11 conditions from 7 evaluation datasets. We lay out a set of recommendations for training the network based on the observed trends. By applying these recommendations to enhance the default recipe provided in the Kaldi toolkit, a significant gain of 13-21% on the Speakers in the Wild and NIST SRE’16 datasets is achieved.

 DOI: 10.21437/Odyssey.2018-46

Cite as: Mclaren, M., Castán, D., Nandwana, M.K., Ferrer, L., Yilmaz, E. (2018) How to train your speaker embeddings extractor . Proc. Odyssey 2018 The Speaker and Language Recognition Workshop, 327-334, DOI: 10.21437/Odyssey.2018-46.

  author={Mitchell Mclaren and Diego Castán and Mahesh Kumar Nandwana and Luciana Ferrer and Emre Yilmaz},
  title={How to train your speaker embeddings extractor	},
  booktitle={Proc. Odyssey 2018 The Speaker and Language Recognition Workshop},