TY - GEN
T1 - Indonesian-English Code-Switching Speech Recognition using the Machine Speech Chain based Semi-Supervised Learning
AU - Tazakka, Rais Vaza Man
AU - Lestari, Dessi
AU - Purwarianti, Ayu
AU - Tanaya, Dipta
AU - Azizah, Kurniawati
AU - Sakti, Sakriani
N1 - Publisher Copyright:
© 2024 ELRA Language Resource Association.
PY - 2024
Y1 - 2024
N2 - Indonesia is home to a diverse linguistic landscape, where individuals seamlessly transition between Indonesian, English, and local dialects in their everyday conversations—a phenomenon known as code-switching. Understanding and accommodating this linguistic fluidity is essential, particularly in the development of accurate speech recognition systems. However, tackling Indonesian-English code-switching poses a challenge due to the scarcity of paired code-switching data. Thus, this study endeavors to address Indonesian-English code-switching in speech recognition, leveraging unlabeled data and employing a semi-supervised technique known as the machine speech chain. Our findings demonstrate that the machine speech chain method effectively enhances automatic speech recognition (ASR) performance in recognizing code-switching between Indonesian and English, utilizing previously untapped resources of unlabeled data.
AB - Indonesia is home to a diverse linguistic landscape, where individuals seamlessly transition between Indonesian, English, and local dialects in their everyday conversations—a phenomenon known as code-switching. Understanding and accommodating this linguistic fluidity is essential, particularly in the development of accurate speech recognition systems. However, tackling Indonesian-English code-switching poses a challenge due to the scarcity of paired code-switching data. Thus, this study endeavors to address Indonesian-English code-switching in speech recognition, leveraging unlabeled data and employing a semi-supervised technique known as the machine speech chain. Our findings demonstrate that the machine speech chain method effectively enhances automatic speech recognition (ASR) performance in recognizing code-switching between Indonesian and English, utilizing previously untapped resources of unlabeled data.
KW - code-switching
KW - machine speech chain
KW - speech recognition systems
UR - http://www.scopus.com/inward/record.url?scp=85195238755&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85195238755
T3 - 3rd Annual Meeting of the ELRA-ISCA Special Interest Group on Under-Resourced Languages, SIGUL 2024 at LREC-COLING 2024 - Workshop Proceedings
SP - 143
EP - 148
BT - 3rd Annual Meeting of the ELRA-ISCA Special Interest Group on Under-Resourced Languages, SIGUL 2024 at LREC-COLING 2024 - Workshop Proceedings
A2 - Melero, Maite
A2 - Sakti, Sakriani
A2 - Soria, Claudia
PB - European Language Resources Association (ELRA)
T2 - 3rd Annual Meeting of the ELRA-ISCA Special Interest Group on Under-Resourced Languages, SIGUL 2024
Y2 - 21 May 2024 through 22 May 2024
ER -