QR-koodi

Multilingual end-to-end ASR for low-resource Turkic languages with common alphabets

Abstract To obtain a reliable and accurate automatic speech recognition (ASR) machine learning model, it is necessary to have sufficient audio data transcribed, for training. Many languages in the world, especially the agglutinative languages of the Turkic family, suffer from a lack of this type of...

Täydet tiedot

Tallennettuna:
Päätekijät: Akbayan Bekarystankyzy, Orken Mamyrbayev, Mateus Mendes, Anar Fazylzhanova, Muhammad Assam
Aineistotyyppi: Artigo
Kieli:Inglês
Julkaistu: Nature Portfolio 2024-06-01
Sarja:Scientific Reports
Aiheet:
End-to-end ASR
Agglutinative languages
Multilingual learning
Conformer
Attention-based
Connectionist temporal classification
Linkit:https://doi.org/10.1038/s41598-024-64848-1
Tagit: Lisää tagi
Ei tageja, Lisää ensimmäinen tagi!

Samankaltaisia teoksia