The texts in the sah column were generated using OCR and may contain errors or artifacts. Please take this into account when using the data for training or evaluation.
The dataset was aligned using the Lingtrain Aligner library (https://github.com/averkij/lingtrain-aligner), created by @averoo
- Downloads last month
- 10