LREC 2000 2nd International Conference on Language Resources & Evaluation
 

Previous Paper   Next Paper

Title Annotation of a Multichannel Noisy Speech Corpus
Authors Cristoforetti L. (ITC-irst (Istituto per la Ricerca Scientifica e Tecnologica) Povo I-38050 Trento, Italy, cristofo@itc.it)
Matassoni M. (ITC-irst (Istituto per la Ricerca Scientifica e Tecnologica) Povo I-38050 Trento, Italy, matasso@itc.it)
Omologo M. (ITC-irst (Istituto per la Ricerca Scientifica e Tecnologica) Povo I-38050 Trento, Italy, omologo,svaizer,zovato@itc.it)
Svaizer P. (ITC-irst (Istituto per la Ricerca Scientifica e Tecnologica) Povo I-38050 Trento, Italy, svaizer@itc.it)
Zovato E. (ITC-irst (Istituto per la Ricerca Scientifica e Tecnologica) Povo I-38050 Trento, Italy, zovato@itc.it)
Keywords Annotation Tools, In-Car Speech Data, JAVA, Multi-Channel Databases, Segmentation
Session Session SP4 - Tools for Evaluation and Processing of Spoken Language Resources
Full Paper 358.ps, 358.pdf
Abstract This paper describes the activity of annotation of an Italian corpus of in-car speech material, with specific reference to the JavaSgram tool, developed with the purpose of annotating multichannel speech corpora. Some pre/post processing tools used with JavaSgram are briefly described together with a synthetic description of the annotation criteria which were adopted. The final objective is that of using the resulting corpus for training and testing a hands-free speech recognizer under development.