Comparison of Self-Supervised Speech Pre-Training Methods on Flemish Dutch

Abstract

Recent research in speech processing exhibits a growing interest in unsupervised and self-supervised representation learning from unlabelled data to alleviate the need for large amounts of annotated data. We investigate several popular pre-training methods and apply them to Flemish Dutch. We compare off-the-shelf English pre-trained models to models trained on an increasing amount of Flemish data. We find that the most important factors for positive transfer to downstream speech recognition tasks include a substantial amount of data and a matching pre-training domain. Ideally, we also finetune on an annotated subset in the target language. All pre-trained models improve linear phone separability in Flemish, but not all methods improve Automatic Speech Recognition. We experience superior performance with wav2vec 2.0 and we obtain a 30% WER improvement by finetuning the multilingually pre-trained XLSR-53 model on Flemish Dutch, after integration into an HMM-DNN acoustic model.

To be published in the 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2021)

References in corpus (7)

• MLS: A Large-Scale Multilingual Dataset for Speech Research
• vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations
• Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition
• An Unsupervised Autoregressive Model for Speech Representation Learning
• Unsupervised Speech Recognition
• Speech SIMCLR: Combining Contrastive and Reconstruction Objective for Self-supervised Speech Representation Learning
• Self-training and Pre-training are Complementary for Speech Recognition