The AMI Meeting Corpus is a multi-modal data set consisting of 100 hours of meeting recordings. It is being created in the context of a project that is developing meeting browsing technology and will eventually be released publicly. Some of the meetings it contains are naturally occurring, and some are elicited, particularly using a scenario in which the participants play different roles in a design team, taking a design project from kick-off to completion over the course of a day. The corpus is being recorded using a wide range of devices including close-talking and far-field microphones, individual and room-view video cameras, projection, a whiteboard, and individual pens, all of which produce output signals that are synchronized with each other. It is also being hand-annotated for many different phenomena, including orthographic transcription, discourse properties such as named entities and dialogue acts, summaries, emotions, and some head and hand gestures. We describe the data set, including the rationale behind using elicited material, and explain how the material is being recorded, transcribed and annotated.

Browsing for elements of interest within a recorded meeting is time-consuming. We describe work in progress on a meeting browser, which aims to support this process by displaying many types of data. These include media, transcripts and processing results, such as speaker segmentations. Users interact with these visualizations to observe and control synchronized playback of the recorded meeting.

No abstract

In this paper we examine the capabilities and limitations of Petrl nets and investigate techniques for proving their correctness.We define different classes of nets where each is basically a Petri net with slight modifications and study the relationship between the various classes.One particular class appears to be quite powerful, with respect to its capability for representing coordinations.In the second part of the paper we establish the feasibility of using the methods of computational induction and inductive assertions to prove restricted statements about Petri nets.

Current solutions for providing access to electronic documents while away from the office do not meet the special needs of mobile document workers. We describe ”Satchel,“ a system that is designed specifically to support the distinctive features of mobile document work. Satchel is designed to meet the following five high-level design goals (1) easy access to document services; (2) timely document access; (3) streamlined user interface; (4) ubiquity; and (5)compliance with security policies. Our current prototype uses a Nokia 9000 Communicator as the mobile device; it communicates to the rest of the Satchel system using wireless communications, both infrared and radio. A fundamental Satchel concept is the use of tokens, or small secure references, to represent documents on the mobile device. The mobile client only transmits small tokens over te wireless channels, leaving the wired network to transmit the contents of documents when, and only when, they are required. Another fundmental Satchel concept is the highly speclized and context-sensitive user interface on the mobile device. The user's interactions ae streamlined because of this specialization and though the use of contextual information gained by uing infrared communications. We report the results of a trial of Satchel that was carried out within our own company, and discuss how well Satchel met our design goals. We call Satchel a ”document appliance” because it provides a streamlined soultion to the problem of remote document access—it aims to support only a limited set of activities, but supports them very well.