Developing State-of-the-Art End-to-End ASR for Norwegian

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F23%3A00012033" target="_blank" >RIV/46747885:24220/23:00012033 - isvavai.cz</a>

Výsledek na webu

<a href="https://dl.acm.org/doi/abs/10.1007/978-3-031-40498-6_18" target="_blank" >https://dl.acm.org/doi/abs/10.1007/978-3-031-40498-6_18</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-40498-6_18" target="_blank" >10.1007/978-3-031-40498-6_18</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Developing State-of-the-Art End-to-End ASR for Norwegian

Popis výsledku v původním jazyce

We present the process of developing a modern end-to-end (E2E) automatic speech recognition (ASR) system for Norwegian (NO), which is a challenging language with many dialects and two written standards (Bokmål and Nynorsk). Since the existing speech corpora for this language are severely limited, we have had to acquire large amounts of additional data. This acquisition has been done by automatic processing of publicly accessible broadcast and parliament archives, YouTube and podcast channels, and also audiobooks. The data-harvesting process has been controlled by the ASR system, whose model has continuously been updated on the extracted chunks of speech. The final model has been trained on 1,246 h of Norwegian and further enhanced by transfer learning from an existing Swedish model. The performance of the ASR system has been evaluated on an 18-h collection of test sets (most of them publicly available) representing different application areas. Our best word error rate (WER) achieved on this collection is 7.6%, which is better than the results obtained from Google and Microsoft cloud services.

Název v anglickém jazyce

Developing State-of-the-Art End-to-End ASR for Norwegian

Popis výsledku anglicky

We present the process of developing a modern end-to-end (E2E) automatic speech recognition (ASR) system for Norwegian (NO), which is a challenging language with many dialects and two written standards (Bokmål and Nynorsk). Since the existing speech corpora for this language are severely limited, we have had to acquire large amounts of additional data. This acquisition has been done by automatic processing of publicly accessible broadcast and parliament archives, YouTube and podcast channels, and also audiobooks. The data-harvesting process has been controlled by the ASR system, whose model has continuously been updated on the extracted chunks of speech. The final model has been trained on 1,246 h of Norwegian and further enhanced by transfer learning from an existing Swedish model. The performance of the ASR system has been evaluated on an 18-h collection of test sets (most of them publicly available) representing different application areas. Our best word error rate (WER) achieved on this collection is 7.6%, which is better than the results obtained from Google and Microsoft cloud services.

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

<a href="/cs/project/TO01000027" target="_blank" >TO01000027: NORDTRANS - Technologie pro automatický přepis řeči ve vybraných severských jazycích</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

Lecture Notes in Computer Science - including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics

ISBN

978-303140497-9

ISSN

03029743

e-ISSN

—

Počet stran výsledku

14

Strana od-do

200-213

Název nakladatele

Springer Science and Business

Místo vydání

—

Místo konání akce

Plzeň, ČR

Datum konání akce

1. 1. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—