Look Who's Not Talking

The objective of this work is speaker diarisation of speech recordings 'in the wild'. The ability to determine speech segments is a crucial part of diarisation systems, accounting for a large proportion of errors. In this paper, we present a simple but effective solution for speech activit...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE Spoken Language Technology Workshop (SLT) pp. 567 - 573
Main Authors Kwon, Youngki, Heo, Hee Soo, Huh, Jaesung, Lee, Bong-Jin, Chung, Joon Son
Format Conference Proceeding
LanguageEnglish
Published IEEE 19.01.2021
Subjects
Conferences
Detectors
speaker diarisation
speaker recognition
speech activity detection
Voice activity detection
Online AccessGet full text

Cover

More Information
Summary:The objective of this work is speaker diarisation of speech recordings 'in the wild'. The ability to determine speech segments is a crucial part of diarisation systems, accounting for a large proportion of errors. In this paper, we present a simple but effective solution for speech activity detection based on the speaker embeddings. In particular, we discover that the norm of the speaker embedding is an extremely effective indicator of speech activity. The method does not require an independent model for speech activity detection, therefore allows speaker diarisation to be performed using a unified representation for both speaker modelling and speech activity detection. We perform a number of experiments on in-house and public datasets, in which our method outperforms popular baselines.
DOI:10.1109/SLT48900.2021.9383502