Skip to Main content Skip to Navigation
Conference papers

Speaker Diarization With Unsupervised Training Framework

Abstract : This paper investigates single and cross-show diarization based on an unsupervised i-vector framework, on French TV and Radio corpora. This framework uses speaker clustering as a way to automatically select data from unlabeled corpora to train i-vector PLDA models. Performances between supervised and unsupervised models are compared. The experimental results on two distinct test corpora (one TV, one Radio) show that unsupervised models perform as good as supervised models for both tasks. Such results indicate that performing an effective cross-show diarization on new language or new domain data in the future should not depend on the availability of manually annotated data.
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01433167
Contributor : Sylvain Meignier <>
Submitted on : Wednesday, March 22, 2017 - 12:18:11 AM
Last modification on : Thursday, April 6, 2017 - 10:15:01 AM
Document(s) archivé(s) le : Friday, June 23, 2017 - 12:34:26 PM

File

speaker-diarization-unsupervis...
Files produced by the author(s)

Identifiers

Collections

Citation

Gaël Le Lan, Sylvain Meignier, Delphine Charlet, Paul Deléglise. Speaker Diarization With Unsupervised Training Framework. 41st IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2016), Mar 2016, Shanghai, China. pp.5, ⟨10.1109/ICASSP.2016.7472741⟩. ⟨hal-01433167⟩

Share

Metrics

Record views

207

Files downloads

899