Evaluate translations via either a self-hosted Embedder or using Chat-GPT.
- (self-hosted) Evaluate translations via a local SBERT embedder
- (self-hosted) Evaluate translations via a local multi-lingual sentence embedder
- (remote) Evaluate translations with LLM as Judge via Chat-GPT
The evaluator can be used as part of a machine learning pipeline, such as EPE (Extract, Predict, Evaluate) - read more on Medium.