Create and query a little Search Index in one language to find top_k similar items in different languages.
E.g. Import a CSV containing Text and URLs
Vectorise text into high dimensional space.
Create an Index we would like to find similar items from
International SEO: Find all semantic similar Pages/URLs in 15+ Languages by searching for one language query.
Legal: Find all semantic similar Paragraphs/Documents in 15+ Langauges by searching for one language query.
Compare: Visualise the semantic similarity of Sentences or Paragraphs within a Corpus.
More? Let me know
Feel free to find an un-optimized example (similar.py) at Github using TF, TFRS and TFhub.
(super-powered by)