{"ID":2867301,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.19270","arxiv_id":"2509.19270","title":"SloPal: A 60-Million-Word Slovak Parliamentary Corpus with Aligned Speech and Fine-Tuned ASR Models","abstract":"Slovak remains a low-resource language for automatic speech recognition (ASR), with fewer than 100 hours of publicly available training data. We present SloPal, a comprehensive Slovak parliamentary corpus comprising 330,000 speaker-segmented transcripts (66 million words, 220 million tokens) spanning 2001--2024, with rich metadata including speaker names, roles, and session information. From this collection, we derive SloPalSpeech, a 2,806-hour aligned speech dataset with segments up to 30 seconds, constructed using a language-agnostic anchor-based alignment pipeline and optimized for Whisper-based ASR training. Fine-tuning Whisper on SloPalSpeech reduces Word Error Rate (WER) by up to 70\\%, with the fine-tuned small model (244M parameters) approaching base large-v3 (1.5B parameters) performance at 6$\\times$ fewer parameters. We publicly release the SloPal text corpus, SloPalSpeech aligned audio, and four fine-tuned Whisper models at https://huggingface.co/collections/NaiveNeuron/slopal, providing the most comprehensive open Slovak parliamentary language resource to date.","short_abstract":"Slovak remains a low-resource language for automatic speech recognition (ASR), with fewer than 100 hours of publicly available training data. We present SloPal, a comprehensive Slovak parliamentary corpus comprising 330,000 speaker-segmented transcripts (66 million words, 220 million tokens) spanning 2001--2024, with r...","url_abs":"https://arxiv.org/abs/2509.19270","url_pdf":"https://arxiv.org/pdf/2509.19270v2","authors":"[\"Erik Božík\",\"Marek Šuppa\"]","published":"2025-09-23T17:33:57Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\",\"cs.SD\"]","methods":"[]","has_code":false}