In this paper, we describe the adaptation of the new BAVIECA ASR engine to Italian. After a brief comparison with other similar systems, we give an overview of the BAVIECA Acoustic and Language Model (AM) training procedure, we describe the final test proce-dures and we end reporting the results obtained so far in some adults and children speech ASR experiments, comparing them with those previously obtained with different recogni-tion engines.
Tipo Pubblicazione:
Contributo in volume
Source:
AISV 2014, 10th Conference of Associazione Italiana di Scienze della Voce, Aspetti prosodici e testuali del raccontare: dalla letteratura orale al parlato dei media, edited by Antonio Romano, Matteo Rivoira, Ilario Meandri, pp. 107–114, 2015
Date:
2015
Resource Identifier:
http://www.cnr.it/prodotto/i/342729
http://www2.pd.istc.cnr.it/pierocosi/papers/PieroCosi/cp-AISV2014.pdf
urn:isbn:978-88-6274-602-1
Language:
Eng