ASR for Romanian Language

Author

Gavat, Inge ; Dumitru, C.O.

Author_Institution

Univ. POLITEHNICA Bucharest, Buchares

fYear

2007

fDate

27-30 June 2007

Firstpage

300

Lastpage

303

Abstract

In this paper we will present progresses made in Automatic Speech Recognition (ASR) for Romanian language. The recognition system is based on statistical modelling with hidden Markov models (HMMs). The progresses concern enhancement of modelling by taking into account the context in form of triphones, improvement of speaker independence by applying a gender specific training and enlargement of the feature categories used to describe speech sequences derived not only from perceptual cepstral analysis but also from perceptual linear prediction.

Keywords

hidden Markov models; natural language processing; speech enhancement; Romanian language; automatic speech recognition; gender specific training; hidden Markov models; perceptual cepstral analysis; perceptual linear prediction; speaker independence; statistical modelling; Acoustic testing; Automatic speech recognition; Context modeling; Feature extraction; Hidden Markov models; Humans; Natural languages; Speech analysis; Speech enhancement; Speech recognition; ASR; HMM; LPC; MFCC; PLP; monophones; triphones;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Signals and Image Processing, 2007 and 6th EURASIP Conference focused on Speech and Image Processing, Multimedia Communications and Services. 14th International Workshop on

Conference_Location

Maribor

Print_ISBN

978-961-248-029-5

Electronic_ISBN

978-961-248-029-5

Type

conf

DOI

10.1109/IWSSIP.2007.4381101

Filename

4381101