Swedish Whispers; Leveraging a Massive Speech Corpus for Swedish Speech Recognition

📅 2025-05-23
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Swedish, as a medium-resource language, suffers from suboptimal performance in multilingual ASR models (e.g., Whisper) due to insufficient representation in training data. To address this, we construct the first high-diversity, cross-domain monolingual Swedish training set—built upon the largest publicly available Swedish speech corpus to date—and systematically perform supervised fine-tuning of the Whisper architecture. Our methodology incorporates rigorous data cleaning and augmentation strategies to significantly enhance model robustness. Evaluation across three major benchmarks—FLEURS, CoVo, and NST—demonstrates that our fine-tuned model achieves an average 47% relative WER reduction over Whisper-large-v3, consistently outperforming the baseline across all model sizes. This work breaks the fine-tuning performance ceiling for medium-resource languages in ASR and establishes a reproducible, high-quality monolingual data-driven paradigm for model optimization.

Technology Category

Application Category

📝 Abstract
This work presents a suite of fine-tuned Whisper models for Swedish, trained on a dataset of unprecedented size and variability for this mid-resourced language. As languages of smaller sizes are often underrepresented in multilingual training datasets, substantial improvements in performance can be achieved by fine-tuning existing multilingual models, as shown in this work. This work reports an overall improvement across model sizes compared to OpenAI's Whisper evaluated on Swedish. Most notably, we report an average 47% reduction in WER comparing our best performing model to OpenAI's whisper-large-v3, in evaluations across FLEURS, Common Voice, and NST.
Problem

Research questions and friction points this paper is trying to address.

Improving Swedish speech recognition accuracy
Fine-tuning Whisper models for mid-resourced languages
Reducing WER by 47% compared to OpenAI's model
Innovation

Methods, ideas, or system contributions that make the work stand out.

Fine-tuned Whisper models for Swedish
Trained on large diverse Swedish dataset
Achieved 47% WER reduction
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
No related jobs found.
L
Leonora Vesterbacka
1KBLab, National Library of Sweden, Sweden
F
Faton Rekathati
1KBLab, National Library of Sweden, Sweden
R
Robin Kurtz
1KBLab, National Library of Sweden, Sweden
J
Justyna Sikora
1KBLab, National Library of Sweden, Sweden
A
Agnes Toftgaard
1KBLab, National Library of Sweden, Sweden