Filipino phoneme recognition trained on the Filipino speech corpus & TIMIT.

This paper describes the experiments and results in the phoneme recognition trained on the Filipino Speech Corpus (FSC) using three sets of training data: TIMIT, FSC and TIMIT plus FSC. Multi-Layer Perceptrons (MLPs) were tarined on these three sets of data. The resulting MLP weights were used to co...

Descrición completa

Detalles Bibliográficos

Publicado en:	UPCOE NEC Technical Bulletin 1, 1 (March 2004).
Autor Principal:	Guevara, Rowena Cristina L.
Formato:	Artigo
Idioma:	English
Subjects:	Bilingual speech recognition. Filipino speech corpus. Multilayer perceptrons. Start-synchronous decoder. Phoneme recognition.