Interactive translation prediction
Workflow for human translation with sentence completion
Interactive translation prediction (ITP) or interactive machine translation (IMT) helps human translators type translations with a predictive text input system. Unlike traditional predictive text, interactive translation prediction uses the machine translation model to provide better completions of the sentence.
User interface from Casmacat
Challenges
- It can be challenging to combine probabilities from the target-side language model with the machine translation model.
See also
- A user study of neural interactive translation prediction (2019) found that interactive translation prediction was more efficient than post-editing with neural machine translation and also highlights user experience challenges.