Where Do Dictionary Apps Actually Get Their Dictionary Data From?

Hi, everyone.
I’ve been looking into dictionary data for a language-learning app I’m building, and I’ve run into something surprising.

There are tons of dictionary apps, browser extensions, reading tools, and language-learning apps that support many languages.

But when I actually try to find good open dictionary data myself, it’s surprisingly difficult.

Even for something as common as English -> Chinese, it’s hard to find a dataset that has all of these:

• headwords
• translations
• part of speech
• multiple senses
• example sentences
• a clear license that allows reuse

I’ve found things like:

• Wiktionary / Wiktextract / Kaikki
• WordNet / Open Multilingual WordNet
• ECDICT for English–Chinese
• CC-CEDICT for Chinese–English
• Tatoeba for bilingual example sentences
• FreeDict
• various language-specific projects

But they all seem to cover different pieces of the puzzle, and the quality and coverage vary quite a bit.

Meanwhile, I regularly see relatively small dictionary apps or browser extensions claiming to support English, Chinese, Japanese, Korean, French, German, Spanish, Italian, Russian, Arabic, etc.

So I’m genuinely curious:
Where does all of that dictionary data usually come from?

I’m especially interested in hearing from anyone who has actually built a dictionary app.

What data sources did you use, and how did you handle licensing?

Btw, I know Cambridge, Oxford, and other major dictionaries offer paid APIs or data licensing, but they’re quite expensive. I assume many smaller dictionary apps probably use their own databases or combine multiple data sources somehow.

Any pointers would be greatly appreciated. Thanks!!!