OCR model for lexical lists in Chinese-IPA Glossing, Ground Truth
Shihua Li, Nathan Wayne Hill · Zenodo (CERN European Organization for Nuclear Research) · 2023
The training was done on 311 pages and validation on 34 pages of ten lexical lists of Burmish languages on Transkribus with the default PyLaia model: (1) Achang • adapted by Hill & Cooper (2020) from Dai & Cui (1985) (2) Bola • adapted by Hill & Cooper (2020) from He & Chen (2004) (3) Bola • adapted by Hill & Cooper (2020) from Dai et al. (2007) (4) Chashan • adapted by Hill & Cooper (2020) from Dai et al. (2010) (5) Langsu • adapted by Hill & Cooper (2020) from He & Chen (2004) (6) Langsu • adapted by Hill & Cooper (2020) from Dai (2005) (7) Leqi • adapted by Hill & Cooper (2020) from He & Chen (2004) (8) Leqi • adapted by Hill & Cooper (2020) from Dai & Li (2006) (9) Leqi • adapted by Hill & Cooper (2020) from Dai & Jie (2007) (10) Zaiwa • adapted by Hill & Cooper (2020) from Xu & Xu (1984)