In this paper, we describe motivations and features of the Speech Recognition task at EVALITA 2009. Systems are compared about the performance on the recognition of uttered connected digits sequences. Interesting results will be shown for various types of approaches, ranging from classic algorithms, to non-standard models. Commercial as well as prototype speech recognizers have been involved into this task and results give an overview of the performance, which leave space to a more general discussion about approaches to speech recognition tasks.
EVALITA 2009: Description and Results of the Speech Recognition task
Gretter, Roberto;Matassoni, Marco
2009-01-01
Abstract
In this paper, we describe motivations and features of the Speech Recognition task at EVALITA 2009. Systems are compared about the performance on the recognition of uttered connected digits sequences. Interesting results will be shown for various types of approaches, ranging from classic algorithms, to non-standard models. Commercial as well as prototype speech recognizers have been involved into this task and results give an overview of the performance, which leave space to a more general discussion about approaches to speech recognition tasks.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
evalita.pdf
non disponibili
Tipologia:
Documento in Post-print
Licenza:
DRM non definito
Dimensione
267.74 kB
Formato
Adobe PDF
|
267.74 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.