Evaluation of Wav2Vec Speech Recognition for Speakers with Cognitive Disorders

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15410%2F22%3A73617758" target="_blank" >RIV/61989592:15410/22:73617758 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-031-16270-1_41" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-16270-1_41</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-16270-1_41" target="_blank" >10.1007/978-3-031-16270-1_41</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Evaluation of Wav2Vec Speech Recognition for Speakers with Cognitive Disorders

Popis výsledku v původním jazyce

. In this paper, we present a spoken dialog system used forcollecting data for future research in the field of dementia predictionfrom speech. The dialog system was used to collect the speech data ofpatients with mild cognitive deficits. The core task solved by the dialogsystem was the spoken description of the vivid shore picture for oneminute. The patients also performed other simple speech-based tasks. Allutterances were recorded and manually transcribed to obtain a groundtruth reference. We describe the architecture of the dialog system as wellas the results of the first speech recognition experiments. The zero-shotWav2Vec 2.0 speech recognizer was used and the recognition accuracyon word- and character-level was evaluated.

Název v anglickém jazyce

Evaluation of Wav2Vec Speech Recognition for Speakers with Cognitive Disorders

Popis výsledku anglicky

. In this paper, we present a spoken dialog system used forcollecting data for future research in the field of dementia predictionfrom speech. The dialog system was used to collect the speech data ofpatients with mild cognitive deficits. The core task solved by the dialogsystem was the spoken description of the vivid shore picture for oneminute. The patients also performed other simple speech-based tasks. Allutterances were recorded and manually transcribed to obtain a groundtruth reference. We describe the architecture of the dialog system as wellas the results of the first speech recognition experiments. The zero-shotWav2Vec 2.0 speech recognizer was used and the recognition accuracyon word- and character-level was evaluated.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

25th International Conference, TSD 2022

ISBN

978-3-031-16270-1

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

12

Strana od-do

"501–512"

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Brno

Datum konání akce

6. 9. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000866222300041