Finding Mnemo is a project dedicated to helping you learn new languages using mnemonics generated by AI models.
It relies on different methods to help you remember tricky words and ideas.
Currently Finding Mnemo is in its prototype version and focuses on providing mnemonics for learning Mandarin words coming from English.
For a given Mandarin word you would want to remember, Finding Mnemo suggests close sounding words in English among a predefined list of candidates.
It then tries to generate a key idea to help remember the pair (original word and matched word).
You are trying to remember the meaning of the word 努力
(effort
).
- Find a homophone / similar sounding word.
The word 努力
is pronounced as nǔ lì
in Mandarin.
A good match for this sound could be the English word newly
.
- Create a mnemonic Key-Phrase.
In order to create a menmonic based on that resemblance, Finding Mnemo suggest a sentence where both words, newly
and effort
, appear.
For example: "It takes effort
to settle in a newly
discovered environment."
- Create a Key-Chain.
An alternative way to remember it is to create a chain of ideas that links both words.
The word 猫咪
is pronounced māomī
. Meaning cat
. The closest-sounding word found in our English vocabulary would be mommy
.
A chain of ideas connecting both concepts could be: cat
⇔ kitten
⇔ mother
⇔ mommy
.
Find a Streamlit demo where you will be able to play with the different functionalities.
Currently, the project can only be installed using its setup.py
file, with pip
.
If you want to learn more, both a user guide and technical documentation can be found in the Wiki Page of the project.