{"ID":5937000,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-09T15:38:11.834581458Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.05224","arxiv_id":"2607.05224","title":"Progressive Refinement: An Iterative Pseudo-Labeling Approach for Mandarin-English Code-Switching ASR","abstract":"Code-switching (CS), alternating languages within the same utterance, poses significant challenges for automatic speech recognition (ASR) due to limited CS training data. This paper applies an iterative pseudo-labeling training approach to CS-ASR for the first time, demonstrating its effectiveness in leveraging unlabeled data to improve CS-ASR performance. The approach comprises three phases: pseudo-label generation, two-stage bilingual model training, and iterative improvements. It begins by generating pseudo-labels from a large unlabeled corpus, creating a semi-supervised dataset. This dataset supports a two-stage training framework where the model is pre-trained and then fine-tuned on supervised CS data. Iterative refinements further enhance the model's accuracy in handling complex CS scenarios. Our approach significantly advances CS-ASR systems, achieving notable Mix Error Rate (MER) reductions on SEAME's devman (6.35%) and devsge (8.29%) subsets.","short_abstract":"Code-switching (CS), alternating languages within the same utterance, poses significant challenges for automatic speech recognition (ASR) due to limited CS training data. This paper applies an iterative pseudo-labeling training approach to CS-ASR for the first time, demonstrating its effectiveness in leveraging unlabel...","url_abs":"https://arxiv.org/abs/2607.05224","url_pdf":"https://arxiv.org/pdf/2607.05224v1","authors":"[\"Qu Yang\",\"Cakra Wardhana\",\"Tim Ng\"]","published":"2026-07-06T15:39:30Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[]","has_code":false}