Toward joint segmentation and classification of dialog acts in multiparty meetings thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Toward joint segmentation and classification of dialog acts in multiparty meetings

Published on Feb 25, 20073036 Views

We present baseline results for the joint segmentation and classification of dialog acts (DAs) of the ICSI Meeting Corpus. Two simple approaches based on word information are investigated and compared

Related categories

Chapter list

Toward Joint Segmentation and classification of Dialog Acts in Multiparty Meetings00:00
ZOrganization00:17
Problem Statement00:26
Problem Statement01:02
Problem Statement01:21
Problem Statement01:37
Problem Statement01:50
Problem Statement02:21
Problem Statement02:44
Hidden Event Language Model03:05
Hidden Event Language Model03:28
Hidden Event Language Model03:31
Hidden Event Language Model03:44
Hidden Event Language Model03:54
Hidden Event Language Model04:05
Tagger04:40
Tagger04:56
Tagger05:00
Tagger05:04
Tagger05:09
Tagger05:50
Performance Metrics06:18
Performance Metrics06:36
Performance Metrics06:48
Performance Metrics06:55
Performance Metrics07:01
Segmentation Metrics07:10
Segmentation Metrics07:33
Segmentation Metrics07:41
Segmentation Metrics08:13
Segmentation Metrics08:47
Segmentation and Classification Metrics09:03
Segmentation and Classification Metrics09:13
Segmentation and Classification Metrics09:28
Segmentation and Classification Metrics09:44
Segmentation and Classification Metrics09:53
Segmentation and Classification Metrics10:10
Segmentation and Classification Metrics10:43
Experimental Setup11:10
Segmentation Performance12:03
Segmentation Performance12:41
Segmentation Performance12:53
Segmentation Performance13:41
Segmentation Performance13:51
Segmentation Performance13:53
Segmentation and Classification Performance13:57
Segmentation and Classification Performance14:15
Segmentation and Classification Performance14:23
Segmentation and Classification Performance14:31
Segmentation and Classification Performance14:45
Segmentation and Classification Performance14:57
Conclusions15:30
Conclusions15:32
Conclusions15:41
Conclusions15:46
Conclusions15:52
Outlook15:57
Outlook15:59
Outlook16:27
Outlook16:39
Thank You16:50