Cyclic Transfer Learning for Mandarin-English Code-Switching Speech Recognition

Nga, Cao Hong; Vu, Duc-Quang; Luong, Huong Hoang; Huang, Chien-Lin; Wang, Jia-Ching

doi:10.1109/lsp.2023.3307350

Cited by 6 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, their large differences at the lexical and syntactic levels allow us to use a single language model (trained on Basque and Spanish texts) that naturally switches from one language to another based only on the acoustic-phonetic clues found by the acoustic models. By the way, this approach makes it easy to deal with code switchings [22] (quite common in Basque Parliament sessions) compared to other recent approaches in the literature [23][24][25][26][27][28][29].…”

Section: Introductionmentioning

confidence: 99%

A Bilingual Basque–Spanish Dataset of Parliamentary Sessions for the Development and Evaluation of Speech Technology

Varona,

Penagarikano,

Bordel

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

The development of speech technology requires large amounts of data to estimate the underlying models. Even when relying on large multilingual pre-trained models, some amount of task-specific data on the target language is needed to fine-tune those models and obtain competitive performance. In this paper, we present a bilingual Basque–Spanish dataset extracted from parliamentary sessions. The dataset is designed to develop and evaluate automatic speech recognition (ASR) systems but can be easily repurposed for other speech-processing tasks (such as speaker or language recognition). The paper first compares the two target languages, emphasizing their similarities at the acoustic-phonetic level, which sets the basis for sharing data and compensating for the relatively small amount of spoken resources available for Basque. Then, Basque Parliament plenary sessions are characterized in terms of organization, topics, speaker turns and the use of the two languages. The paper continues with the description of the data collection procedure (involving both speech and text), the audio formats and conversions along with the creation and postprocessing of text transcriptions and session minutes. Then, it describes the semi-supervised iterative procedure used to cut, rank and select the training segments and the manual supervision process employed to produce the test set. Finally, ASR experiments are presented using state-of-the-art technology to validate the dataset and to set a reference for future works. The datasets, along with models and recipes to reproduce the experiments reported in the paper, are released through Hugging Face.

show abstract