White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Using Deep Neural Networks for Speaker Diarisation

Milner, Rosanna Margaret (2016) Using Deep Neural Networks for Speaker Diarisation. PhD thesis, University of Sheffield.

[img]
Preview
Text
final_thesis.pdf
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (4Mb) | Preview

Abstract

Speaker diarisation answers the question “who spoke when?” in an audio recording. The input may vary, but a system is required to output speaker labelled segments in time. Typical stages are Speech Activity Detection (SAD), speaker segmentation and speaker clustering. Early research focussed on Conversational Telephone Speech (CTS) and Broadcast News (BN) domains before the direction shifted to meetings and, more recently, broadcast media. The British Broadcasting Corporation (BBC) supplied data through the Multi-Genre Broadcast (MGB) Challenge in 2015 which showed the difficulties speaker diarisation systems have on broadcast media data. Diarisation is typically an unsupervised task which does not use auxiliary data or information to enhance a system. However, methods which do involve supplementary data have shown promise. Five semi-supervised methods are investigated which use a combination of inputs: different channel types and transcripts. The methods involve Deep Neural Networks (DNNs) for SAD, DNNs trained for channel detection, transcript alignment, and combinations of these approaches. However, the methods are only applicable when datasets contain the required inputs. Therefore, a method involving a pretrained Speaker Separation Deep Neural Network (ssDNN) is investigated which is applicable to every dataset. This technique performs speaker clustering and speaker segmentation using DNNs successfully for meeting data and with mixed results for broadcast media. The task of diarisation focuses on two aspects: accurate segments and speaker labels. The Diarisation Error Rate (DER) does not evaluate the segmentation quality as it does not measure the number of correctly detected segments. Other metrics exist, such as boundary and purity measures, but these also mask the segmentation quality. An alternative metric is presented based on the F-measure which considers the number of hypothesis segments correctly matched to reference segments. A deeper insight into the segment quality is shown through this metric.

Item Type: Thesis (PhD)
Academic Units: The University of Sheffield > Faculty of Engineering (Sheffield) > Computer Science (Sheffield)
The University of Sheffield > Faculty of Science (Sheffield) > Computer Science (Sheffield)
Depositing User: Rosanna Margaret Milner
Date Deposited: 17 Mar 2017 14:41
Last Modified: 17 Mar 2017 14:41
URI: http://etheses.whiterose.ac.uk/id/eprint/16567

Actions (repository staff only: login required)