Hearing voices at the National Library -- a speech corpus and acoustic model for the Swedish language
This work addresses the need for better speech-to-text technology for Swedish, particularly for cultural heritage institutions with large unlabeled audiovisual collections, though it is incremental as it builds on existing architectures.
The researchers tackled the problem of automated speech recognition for Swedish by developing new acoustic models using wav2vec 2.0 and speech corpora from the National Library of Sweden, resulting in the VoxRex model that outperforms existing Swedish ASR models and shows improved performance when combined with pretrained language models.
This paper explains our work in developing new acoustic models for automated speech recognition (ASR) at KBLab, the infrastructure for data-driven research at the National Library of Sweden (KB). We evaluate different approaches for a viable speech-to-text pipeline for audiovisual resources in Swedish, using the wav2vec 2.0 architecture in combination with speech corpuses created from KB's collections. These approaches include pretraining an acoustic model for Swedish from the ground up, and fine-tuning existing monolingual and multilingual models. The collections-based corpuses we use have been sampled from millions of hours of speech, with a conscious attempt to balance regional dialects to produce a more representative, and thus more democratic, model. The acoustic model this enabled, "VoxRex", outperforms existing models for Swedish ASR. We also evaluate combining this model with various pretrained language models, which further enhanced performance. We conclude by highlighting the potential of such technology for cultural heritage institutions with vast collections of previously unlabelled audiovisual data. Our models are released for further exploration and research here: https://huggingface.co/KBLab.