Back to Search Start Over

Designing an Effective Metric Learning Pipeline for Speaker Diarization

Authors :
Narayanaswamy, Vivek Sivaraman
Thiagarajan, Jayaraman J.
Song, Huan
Spanias, Andreas
Publication Year :
2018

Abstract

State-of-the-art speaker diarization systems utilize knowledge from external data, in the form of a pre-trained distance metric, to effectively determine relative speaker identities to unseen data. However, much of recent focus has been on choosing the appropriate feature extractor, ranging from pre-trained $i-$vectors to representations learned via different sequence modeling architectures (e.g. 1D-CNNs, LSTMs, attention models), while adopting off-the-shelf metric learning solutions. In this paper, we argue that, regardless of the feature extractor, it is crucial to carefully design a metric learning pipeline, namely the loss function, the sampling strategy and the discrimnative margin parameter, for building robust diarization systems. Furthermore, we propose to adopt a fine-grained validation process to obtain a comprehensive evaluation of the generalization power of metric learning pipelines. To this end, we measure diarization performance across different language speakers, and variations in the number of speakers in a recording. Using empirical studies, we provide interesting insights into the effectiveness of different design choices and make recommendations.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.1811.00183
Document Type :
Working Paper