Title :
Audio information access from meeting rooms
Author :
Renals, Steve ; Ellis, Dan
Author_Institution :
Dept. of Comput. Sci., Univ. of Sheffield, UK
Abstract :
We investigate approaches to accessing information from the streams of audio data that result from multi-channel recordings of meetings. The methods investigated use word-level transcriptions, and information derived from models of speaker activity and speaker turn patterns. Our experiments include spoken document retrieval for meetings, automatic structuring of meetings based on self-similarity matrices of speaker turn patterns and a simple model of speaker activity. Meeting recordings are rich in both lexical and non-lexical information; our results illustrate some novel kinds of analysis made possible by a transcribed corpus of natural meetings.
Keywords :
audio signal processing; document handling; information retrieval; matrix algebra; natural languages; speech processing; speech recognition; audio information access; automatic meeting structuring; lexical information; meeting rooms; multi-channel recordings; natural meetings; nonlexical information; self-similarity matrices; speaker activity; speaker turn patterns; speech recognizer output; spoken document retrieval; word-level transcriptions; Audio recording; Automatic speech recognition; Broadcasting; Computer science; Indexing; Information analysis; Information retrieval; Microphones; Speech recognition; Streaming media;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on
Print_ISBN :
0-7803-7663-3
DOI :
10.1109/ICASSP.2003.1202750