Whispering in Norwegian: Navigating Orthographic and Dialectic Challenges

This paper presents NB-Whisper, a tailored adaptation of OpenAI’s Whisper model, specifically fine-tuned to address the unique challenges of Norwegian language Automatic Speech Recognition (ASR). We highlight its key contributions and summarise the results achieved in converting spoken Norwegian into written forms and translating other languages into Norwegian. By training on a 22,000 hour weakly aligned dataset, we show that we are able to improve the Norwegian Bokmål transcription by OpenAI Whisper Large-v3 from a WER of 10.4 to 6.6 on the Fleurs Dataset and from 6.8 to 2.2 on the NST dataset.