Who spoke when: Difference between revisions
From HLT@INESC-ID
No edit summary |
mNo edit summary |
||
(2 intermediate revisions by 2 users not shown) | |||
Line 1: | Line 1: | ||
__NOTOC__ | __NOTOC__ | ||
{| align="right" cellpadding="3" style='width: 35%; align: right; border-style: solid; border-width: 1px; background: #dee2ff; margin-left: 8px;' | {| align="right" cellpadding="3" style='width: 35%; align: right; border-style: solid; border-width: 1px; background: #dee2ff; margin-left: 8px;' | ||
! style='text-align: center;' | | ! style='text-align: center;' |Janez Žibert | ||
|- | |- | ||
| Janez | | Janez Žibert just finished his PhD on the problem of structuring the audio data in terms of speakers. | ||
|} | |} | ||
== Date == | == Date == | ||
* 15:00, November | * 15:00, November 23, 2006 | ||
* | * 4th floor meeting room | ||
== Speaker == | == Speaker == | ||
* [http://luks.fe.uni-lj.si/en/ Janez | * [http://luks.fe.uni-lj.si/en/staff/janezz/index.html Janez Žibert] | ||
== Abstract == | == Abstract == | ||
The thesis addresses the problem of structuring the audio data in | The thesis addresses the problem of structuring the audio data in terms of speakers, i.e., finding the regions in the audio streams that belong to one speaker and joining each region of the same speaker together. The task of organizing the audio data in this way is known as speaker diarization and was first introduced in the NIST project of Rich Transcription in "Who spoke when" evaluations. The speaker-diarization problem is composed of several tasks. This thesis addresses three of them: speech/non-speech segmentation, speaker- and background-change detection, and speaker clustering. | ||
terms of speakers, i.e., finding the regions in the audio streams | |||
that belong to one speaker and joining each region of the same | |||
speaker together. The task of organizing the audio data in this way | |||
is known as speaker diarization and was first introduced in the NIST | |||
project of Rich Transcription in "Who spoke when" | |||
evaluations. The speaker-diarization problem is composed of several | |||
tasks. This thesis addresses three of them: speech/non-speech | |||
segmentation, speaker- and background-change detection, and speaker | |||
clustering | |||
The main objectives in our research were to develop new representations of audio data that were more suitable for each task and to improve the accuracy and increase the robustness of standard approaches under various acoustic and environmental conditions. The motivation for the improvement of the existing methods and the development of new procedures for speaker-diarization tasks is the design of a system for the speaker-based audio indexing of broadcast news shows. | |||
[[category:Seminars]] | [[category:Seminars]] | ||
[[category:Seminars 2006]] | [[category:Seminars 2006]] |
Latest revision as of 00:16, 24 November 2006
Janez Žibert |
---|
Janez Žibert just finished his PhD on the problem of structuring the audio data in terms of speakers. |
Date
- 15:00, November 23, 2006
- 4th floor meeting room
Speaker
Abstract
The thesis addresses the problem of structuring the audio data in terms of speakers, i.e., finding the regions in the audio streams that belong to one speaker and joining each region of the same speaker together. The task of organizing the audio data in this way is known as speaker diarization and was first introduced in the NIST project of Rich Transcription in "Who spoke when" evaluations. The speaker-diarization problem is composed of several tasks. This thesis addresses three of them: speech/non-speech segmentation, speaker- and background-change detection, and speaker clustering.
The main objectives in our research were to develop new representations of audio data that were more suitable for each task and to improve the accuracy and increase the robustness of standard approaches under various acoustic and environmental conditions. The motivation for the improvement of the existing methods and the development of new procedures for speaker-diarization tasks is the design of a system for the speaker-based audio indexing of broadcast news shows.