… Sámi alphabet, and our OCR models could improve upon NLN’s transcription for Inari Sámi.
… and are good candidates for a re-OCR process. If transcription accuracy is the main focus, …

… The GT4HistOCR dataset is particularly well-suited for training advanced recognition models
in OCR software that utilize recurrent neural networks, specifically the LSTM architecture, …

… Optical Character Recognition (henceforth OCR) tools are … more systematic comparisons of
different OCR tools available; see… This is also a very particular case for OCR, since the scripts …

… In our own projects, we have developed language technologies focusing on low-resource
scenarios, specifically for the two Uralic languages Zyrian Komi and North Saami. In addition …

… to contain errors introduced by OCR (optical character recognition) methods used in the …
to-sequence NMT (neural machine translation) model to conduct OCR error correction designed …

… , neural models, language documentation, tokenisation, corpora and lexicons, optical character
recognition, … , Hungarian and Estonian as well as North Sámi, Livonian and Votic, and …

… To reach an end-user suitable quality of the TTS, we have used a neural, end-toend … ),
which was scanned and OCR-processed for the project. Additional usable material came …

… versus neural networks. Why would you write rules for an endangered language if neural …
Multilingual dependency parsing for lowresource languages: Case studies on north saami …

… indeed, has had only one fourth of the exposure to the Kildin Saami special characters that
the monolingual Kildin Saami model received. Nevertheless, the results are very close. Even …

… workplace software, and optical character recognition (OCR) … —Estonian, North Sámi, and
South Sámi— spelling … test set, this work shows that neurological embeddings can be used to …