Detecting English Speech in the Air Traffic Control Voice Communication

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F21%3APU142967" target="_blank" >RIV/00216305:26230/21:PU142967 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.isca-speech.org/archive/interspeech_2021/szoke21_interspeech.html" target="_blank" >https://www.isca-speech.org/archive/interspeech_2021/szoke21_interspeech.html</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.21437/Interspeech.2021-1033" target="_blank" >10.21437/Interspeech.2021-1033</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Detecting English Speech in the Air Traffic Control Voice Communication

Popis výsledku v původním jazyce

Developing in-cockpit voice enabled applications require a realworld dataset with labels and annotations. We launched a community platform for collecting the Air-Traffic Control (ATC) speech, world-wide in the ATCO2 project. Filtering out non- English speech is one of the main components in the data processing pipeline. The proposed English Language Detection (ELD) system is based on the embeddings from Bayesian subspace multinomial model. It is trained on the word confusion network from an ASR system. It is robust, easy to train, and light weighted. We achieved 0:0439 equal-error-rate (EER), a 50% relative reduction as compared to the state-of-the-art acoustic ELD system based on x-vectors, in the in-domain scenario. Further, we achieved an EER of 0:1352, a 33% relative reduction as compared to the acoustic ELD, in the unseen language (out-of-domain) condition. We plan to publish the evaluation dataset from the ATCO2 project.

Název v anglickém jazyce

Detecting English Speech in the Air Traffic Control Voice Communication

Popis výsledku anglicky

Developing in-cockpit voice enabled applications require a realworld dataset with labels and annotations. We launched a community platform for collecting the Air-Traffic Control (ATC) speech, world-wide in the ATCO2 project. Filtering out non- English speech is one of the main components in the data processing pipeline. The proposed English Language Detection (ELD) system is based on the embeddings from Bayesian subspace multinomial model. It is trained on the word confusion network from an ASR system. It is robust, easy to train, and light weighted. We achieved 0:0439 equal-error-rate (EER), a 50% relative reduction as compared to the state-of-the-art acoustic ELD system based on x-vectors, in the in-domain scenario. Further, we achieved an EER of 0:1352, a 33% relative reduction as compared to the acoustic ELD, in the unseen language (out-of-domain) condition. We plan to publish the evaluation dataset from the ATCO2 project.

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

2021

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings Interspeech 2021

ISBN

—

ISSN

1990-9772

e-ISSN

—

Počet stran výsledku

5

Strana od-do

3286-3290

Název nakladatele

International Speech Communication Association

Místo vydání

Brno

Místo konání akce

Brno

Datum konání akce

30. 8. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—