Analyzer with Automatic language detection
Our use case is that users upload files which can be in any languages, hence we have to do language identification on the content. Each file only exists in one language, meaning we don't have multiple translations of the same file.
For now that means we have to create one field per potential language in the index and assign them corresponding language analyzers. Then when we index the file we upload it's content to the field matching the language. Then when we add more field with the same language requirements we end up with spending a huge number of fields just to manage languages...
What I had hoped to se was an "auto" mode for language analyzers. So that instead of us having to identify the language of the content and select the corresponding index field to put it in, the search engine could manage this internally.
You can accomplish this with a skillset and a custom language detection skill. The documentation for creating a skillset can be found here https://docs.microsoft.com/en-us/azure/search/cognitive-search-defining-skillset
Azure Search Product Team