Hi, I want to extract token translations from aligned sentences (DE - EN) with prodigy. Can you give me some advice, which annotation recipe(s) to use best in this case? In the end, I would like to obtain a trained model, that can predict the token translation pairs of given sentence alignments (based on the annotations). Thanks in advance for any help!
Hi! Just to make sure I understand the exact use case and annotation task: You have two aligned sentences and you now want to match up the exact tokens? Or do already have specific spans of text in sentence A that you now want to highlight in sentence B?
Also, do you already have a model you want to use for this task? Are there existing predictions you want to include, or are you just looking to create data that you can then export for training?