March 2022 – The Nyingarn Platform workspace has been built and the Project Team are now testing ways to convert images of manuscripts to text using either Optical Character Recognition (OCR) or crowdsourced transcription platforms. We have tested two OCR systems with differing results for typed and handwritten manuscripts.
Manuscripts with existing permissions have also been transcribed through the online crowdsourcing platform, DigiVol, hosted by the Australian Museum. We have created manuscript transcription ‘expeditions’ in DigiVol that allow us to refine transcribing guides for volunteers and then export this language data into our workspace. To date, expeditions of 20-40 pages have been transcribed by volunteers within days. At Nyingarn, we will continue to refine both these transcription techniques as more manuscripts in state and national institutions are identified by CIs and Steering Committee members.
Registered users can access the workspace