Robust ASR front-end using spectral-based and discriminant features: experiments on Aurora tasks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F01%3APU23547" target="_blank" >RIV/00216305:26220/01:PU23547 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Robust ASR front-end using spectral-based and discriminant features: experiments on Aurora tasks

Popis výsledku v původním jazyce

This paper describes an automatic speech recognition front-end that combines low-level robust ASR feature extraction tech-niques, and higher-level linear and non-linear feature transformations. The low-level algorithms use data-derived filters, mean andvariance normalization of the feature vectors, and dropping of noise frames. The feature vectors are then linearly transformed using Principal Components Analysis (PCA). An Artificial Neural Network (ANN) is also used to compute features that are usefulfor classification of speech sounds. It is trained for phoneme probability estimation on a large corpus of noisy speech. These transformations lead to two feature streams whose vectors are concatenated and then used for speech recognition. This method was tested on the set of speech corpora used for the "Aurora" evaluation. Using the feature stream generated without the ANN yields an overall 41% reduction of the error rate over Mel-Frequency Cepstral Coefficients (MFCC) reference feature

Název v anglickém jazyce

Robust ASR front-end using spectral-based and discriminant features: experiments on Aurora tasks

Popis výsledku anglicky

This paper describes an automatic speech recognition front-end that combines low-level robust ASR feature extraction tech-niques, and higher-level linear and non-linear feature transformations. The low-level algorithms use data-derived filters, mean andvariance normalization of the feature vectors, and dropping of noise frames. The feature vectors are then linearly transformed using Principal Components Analysis (PCA). An Artificial Neural Network (ANN) is also used to compute features that are usefulfor classification of speech sounds. It is trained for phoneme probability estimation on a large corpus of noisy speech. These transformations lead to two feature streams whose vectors are concatenated and then used for speech recognition. This method was tested on the set of speech corpora used for the "Aurora" evaluation. Using the feature stream generated without the ANN yields an overall 41% reduction of the error rate over Mel-Frequency Cepstral Coefficients (MFCC) reference feature

Druh

D - Stať ve sborníku

CEP obor

JA - Elektronika a optoelektronika, elektrotechnika

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2001

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proc. EUROSPEECH

ISBN

87-90834-09-7

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

429-432

Název nakladatele

Neuveden

Místo vydání

Aalborg

Místo konání akce

Aalborg

Datum konání akce

3. 9. 2001

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—