Get Word - An app for studying languages
A small sample from the app. Just scratch off the covered translation and choose how soon the word should come back.
The most frequent words keep coming back, which is why coverage grows quickly at the start and the last few percent come slowest.
The estimate comes from a measured curve for spoken English: the first 100 most frequent words cover around 64% of everyday speech, 500 words around 84%, 1 000 words around 89% and 3 000 words around 95%.
The last few percent are worth more than they look. The difference between 95% and 98% coverage is the difference between one unknown word in twenty and one in fifty, more than twice as few. And the fewer unknown words you have around you, the easier it is to work out what they mean from what surrounds them.
Coverage here means overall coverage of the language by word frequency, from the most frequent words down to the rarest. Coverage is not the same as understanding, though. It estimates how often you run into a word you do not know, not how much of what was said you end up following. Understanding is also shaped by grammar, by how fast people speak and by how well you know the topic. On a familiar topic much lower overall coverage is enough, while on an unfamiliar one even high coverage can leave noticeable gaps.
SourcesAdolphs & Schmitt (2003): Lexical Coverage of Spoken Discourse (abstract)Nation (2006): How Large a Vocabulary Is Needed for Reading and Listening?Nation & Waring (1997): Vocabulary Size, Text Coverage and Word Lists
Thanks to spaced repetition you will not completely forget what you have already learned, or needlessly repeat what you already know.
There is not much point in learning words or phrases you will use once in five years, or already know well, and skipping the ones you use almost every day. This is quite different for everyone, we all have a slightly different job and different hobbies. That is why the selection follows what you decide.
You can add words: from a photo, by typing, with an AI bot, or from a conversation (we are working on that one).
From a photo you upload, the AI works out what is what, makes the translations, generates the audio, and you then pick which words you want to learn.
You just write what you want to be able to say in the other language → a translation is created, audio is generated, and the word or sentence is added to your vocabulary.
When you tell the AI bot what you would like to be able to say, it prepares a list you can then pick from — whatever you need or want.
This one is in development. I would go as far as saying it can make adding new words incredibly easy.
We support combinations of any two languages. To be completely precise, for about 99% of the languages that are spoken. Ancient languages such as Old Persian or Babylonian are not there right now, but do get in touch if you would be interested in something like that and we will look into it.
We are an open-source app. We will be glad for a contribution to the code, a star on GitHub, or a reported problem.
github.com/jan-miksik/get-wordMini bonus: double-click the Get Word logo and a small surprise is waiting for you :)