Abstract
In this paper we discuss the performance analysis of a speaker diarization system similar to the system that was submitted by ICSI at the NIST RT06s evaluation benchmark. The analysis that is based on a series of oracle experiments, provides a good understanding of the performance of each system component on a test set of twelve conference meetings used in previous NIST benchmarks. Our analysis shows that the speech activity detection component contributes most to the total diarization error rate (23%). The lack of ability to model verlapping speech is also a large source of errors (22%) followed by the component that creates the initial system models (15%).
Original language | English |
---|---|
Title of host publication | Proceedings of Interspeech 2007 |
Place of Publication | Antwerp |
Publisher | International Speech Communication Association (ISCA) |
Pages | ThC.O1-3 |
Number of pages | 4 |
ISBN (Print) | 1990-9772 |
Publication status | Published - 27 Aug 2007 |
Event | 8th Annual Conference of the International Speech Communication Association, INTERSPEECH 2007 - Antwerp, Belgium Duration: 27 Aug 2007 → 31 Aug 2007 Conference number: 8 https://www.interspeech2007.org/ |
Publication series
Name | |
---|---|
Publisher | International Speech Communication Association |
Number | LNCS4549 |
ISSN (Print) | 1990-9772 |
Conference
Conference | 8th Annual Conference of the International Speech Communication Association, INTERSPEECH 2007 |
---|---|
Abbreviated title | INTERSPEECH |
Country/Territory | Belgium |
City | Antwerp |
Period | 27/08/07 → 31/08/07 |
Internet address |
Keywords
- IR-64328
- Speaker diarization
- METIS-241880
- EC Grant Agreement nr.: FP6/506811
- EWI-11002
- rich transcription