High Performance Italian Continuous "Digit" recognition

The development of a speaker independent connected "digits" recognizer for Italian is described. The CSLU Speech Toolkit was used to develop and implement the system which is based on an hybrid ANN/HMM architecture. The recognizer is trained on contextdependent categories to account for coarticulatory variation. Various front-end processing was compared and, when the best features (MFCC with CMS + ?) were considered, there was a 98.68% word recognition accuracy (90.76% sentence recognition accuracy) on a test set of the FIELD continuous digits recognition task.

Tipo Pubblicazione: 
Contributo in volume
Author or Creator: 
Cosi P.
Hosom J.P.
Tesser F.
Publisher: 
China Military Friendship Publish, Beijing, CHN
Source: 
ICSLP 2000 - 6th International Conference on Spoken language Processing, edited by Baozong Yuan, Taiyi Huang, Xiaofang Tang, pp. 242–245. Beijing: China Military Friendship Publish, 2000
Date: 
2000
Resource Identifier: 
http://www.cnr.it/prodotto/i/240951
http://www2.pd.istc.cnr.it/Papers/PieroCosi/cp-ICSLP2000-01.pdf
urn:isbn:7-80150-114-4
Language: 
Eng
ISTC Author: 
Ritratto di Fabio Tesser
Real name: 
Ritratto di Piero Cosi
Real name: