In this work recognition of children`s speech was investigated by considering a phone recognition task. Two baseline systems were trained, one for childrenand one for adults, by exploiting two Italian speech databases. Under matching conditions, training and recognition performed with data from the same population group, the phone recognition accuracy was 77.30% and 79.43% for children and adults, respectively. It was found that for many children recognition results were as good as for adults. However, for children an higher variability in phone recognition accuracy across speakers was observed, than for adults. Vocal tract length normalization, under matched and mismatched training and testing conditions, was also investigated. For both adults and children a performance improvement, with respect to the baseline systems, was observed.
Investigating Recognition of Children`s Speech
Giuliani, Diego;Gerosa, Matteo
2003-01-01
Abstract
In this work recognition of children`s speech was investigated by considering a phone recognition task. Two baseline systems were trained, one for childrenand one for adults, by exploiting two Italian speech databases. Under matching conditions, training and recognition performed with data from the same population group, the phone recognition accuracy was 77.30% and 79.43% for children and adults, respectively. It was found that for many children recognition results were as good as for adults. However, for children an higher variability in phone recognition accuracy across speakers was observed, than for adults. Vocal tract length normalization, under matched and mismatched training and testing conditions, was also investigated. For both adults and children a performance improvement, with respect to the baseline systems, was observed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
