English

RO-N3WS: Enhancing Generalization in Low-Resource ASR with Diverse Romanian Speech Benchmarks

Computation and Language 2026-03-04 v1 Machine Learning Sound

Abstract

We introduce RO-N3WS, a benchmark Romanian speech dataset designed to improve generalization in automatic speech recognition (ASR), particularly in low-resource and out-of-distribution (OOD) conditions. RO-N3WS comprises over 126 hours of transcribed audio collected from broadcast news, literary audiobooks, film dialogue, children's stories, and conversational podcast speech. This diversity enables robust training and fine-tuning across stylistically distinct domains. We evaluate several state-of-the-art ASR systems (Whisper, Wav2Vec 2.0) in both zero-shot and fine-tuned settings, and conduct controlled comparisons using synthetic data generated with expressive TTS models. Our results show that even limited fine-tuning on real speech from RO-N3WS yields substantial WER improvements over zero-shot baselines. We will release all models, scripts, and data splits to support reproducible research in multilingual ASR, domain adaptation, and lightweight deployment.

Keywords

Cite

@article{arxiv.2603.02368,
  title  = {RO-N3WS: Enhancing Generalization in Low-Resource ASR with Diverse Romanian Speech Benchmarks},
  author = {Alexandra Diaconu and Mădălina Vînaga and Bogdan Alexe},
  journal= {arXiv preprint arXiv:2603.02368},
  year   = {2026}
}