Actions

Multiple-languages Language Multiple-languages Culture: «Ideas about language learning»

From Polyglot Club WIKI

< Language‎ | Multiple-languages‎ | Culture
This lesson can still be improved. EDIT IT NOW! & become VIP
Rate this lesson:
5.00
(2 votes)

Interesting ideas about language learning.jpg

This is for expressing your ideas about language learning and culture exchange.

Everyone may say something under its username. And generally, do not delete other users' content.

Contents

GrimPixel[edit | edit source]

GrimPixel has now conceived two major routes:

  1. Gather vocabulary databases free for non-commercial use -> develop a program to unify databases to TSV files -> develop programs to learn them efficiently;
  2. Gather grammatical rules -> develop programs to practice.

He personally prefers to use NodeGui to develop and the license of the programs be MIT. He has started learning related languages (HTML, CSS, JavaScript). Any help is welcomed.

Website of visual dictionary[edit | edit source]

I have seen three books:

5 Language Visual Dictionary

The Firefly Five Language Visual Dictionary: English, Spanish, French, German, Italian

Merriam-Webster's Compact 5-Language Visual Dictionary (English, Spanish, French, German and Italian Edition)

So I think there can be a website that provides such a service, even better.

Prerequisites: word lists, translations, images.

[REALIZED: https://babadada.com/ also some paid visual dictionaries in apple app store]

Review: In Babadada, there are no genders, no pinyin tones for Chinese, no furigana or romaji for Japanese, many languages are incomplete without indications, parts of speech are not clear. Great for having dialects.

There are several visual dictionaries in the apple app store, but they are not for advanced users.

Tutorials of languages[edit | edit source]

I have seen two channels on Youtube:

https://www.youtube.com/user/ProgrammingKnowledge

https://www.youtube.com/channel/UCfx2dro_w4_MyA19Nm5badg

providing "tutorials" about programming languages such as C++, Java, Python, which last several hours. So I think somebody may give such "tutorials" about the grammar of human languages.

Prerequisites: extensive knowledge of a language, language teaching knowledge.

Filter of characteristics[edit | edit source]

This requires a database of characteristics of ethnics or nationalities, with a filter. Input the characteristics, and get the result of which ethnics or nationalities possess them.

Of course, it requires the update, because culture may change with time.

Prerequisites: extensive knowledge of cultures.

Sentence analyzer[edit | edit source]

Tatoeba has a lot of sentences, but it only shows what a sentence looks like. There should be something radical to reveal the function of a sentence and to show people the logicality of a sentence. That is to say, syntax.

I imagine a service, which draws parse trees of each sentence, and shows how a sentence may be changed in the structure into another one without changing its general meaning.

If you don't know how to analyze a sentence, see https://www.wikihow.com/Analyze-Sentences.

Prerequisites: programming knowledge, syntax knowledge.

[REALIZED:

https://www.webforditas.hu/parser

http://www.link.cs.cmu.edu/link/submit-sentence-4.html]

Review: It is clear that linguistics have this tool since long time ago. But it is really not widespread, because they are usually amateur-proof. I wish there will be support for more languages.

Parallel text in video games and other software, also websites[edit | edit source]

It is very common for current games to be multilingual, but no one has made such a little change to help language-learning players. Also in other software, and also websites.

There seems to be a practical way for websites: user script. Each script adds a language on the website.

For compactness, it is possible that texts are not side-by-side, but only in one language, hotkey can be used to alternate languages.

Prerequisites: programming knowledge.

Postage stamps on e-mails[edit | edit source]

Postage stamps can be brought back, showing different culture.

Prerequisites: programming knowledge.

Screensaver for language learning[edit | edit source]

There is an XScreenSaver on Linux. Somebody can make use of it to learn languages, just like "Molecule" teaching chemistry.

Prerequisites: programming knowledge.

Software on the resources[edit | edit source]

This would be epic.

I am thinking about what it can do:

  1. Fetch Wiktionary word frequency lists, translate those words with those dictionaries, generate flashcards on those flashcard programs with example sentences;
  2. Access almost every resource on the lists, with Android or iOS simulator;
  3. Background music from "Traditional Music Channel";

I think it can be named "Polyglot 1".

Prerequisites: programming knowledge.

Google Maps game[edit | edit source]

It shows a random street view image, and let you find its origin in a range of satellite images by switching to street view and reaching the same spot.

Prerequisites: programming knowledge.

[SIMILAR: https://www.geoguessr.com/]

Review on Geoguessr: It is not what I want. My idea is based on how to locate a photo, but this one just wants to let you know where you are. I think its idea is not so cool as mine.

A remix of Tetris Theme A played by instruments of the world[edit | edit source]

This song is very good for looping. So I think it may be remixed involving all kinds of instruments around the world, each one plays a section.

Prerequisites: players.

Make use of Applets, Widgets, Gadgets, etc.[edit | edit source]

It is always forgotten that operating systems have these stuff. Applets on Debian, Widgets on Android, Gadgets on Windows, other names on other platforms. They are highly undervalued. They can start automatically right after boot. I can be forced to see them, in case I were too busy to remember to learn languages.

Vocabulary on loading screen[edit | edit source]

When a program is loading, a word and its translation show up on the screen. It may be a mod for a game, or an add-on for a browser.

Prerequisites: programming knowledge, word lists.

[REALIZED: https://mainichi.me/]

Etymology trees of words[edit | edit source]

Just like family trees of the human, words have their trees.

I don't mean “word family”, I mean that a word root that cannot be retrieved further is the ancestor on the “family tree”, and the words derived from it are the branches. Precisely, different forms of the same lexeme should be put on the z-axis and only be displayed with “expand” button pressed. The lemma represents the other forms on the tree.

The trees are cross-language, showing lexical relations between languages.

Prerequisites: programming knowledge.

Altering TTS[edit | edit source]

Read a word or a sentence in a language in a line, automatically change the voice in the next line. This will help creating recordings of vocabulary or dialogue.

Prerequisites: programming knowledge.

Mods for learning[edit | edit source]

Create mods in games, to allow players learn languages and other subjects.

The mods can be books to read, recordings to listen to, and so on. Just like learning a language in the real life.

Prerequisites: programming knowledge.

[REALIZED:

https://www.nexusmods.com/skyrim/mods/8892

https://www.nexusmods.com/skyrim/mods/46088

https://www.nexusmods.com/skyrim/mods/24456

https://www.nexusmods.com/skyrim/mods/76477

https://www.nexusmods.com/skyrim/mods/57436

https://www.nexusmods.com/skyrim/mods/94931]

Review: Well, how I wish people can work on an open-source game like this.

Game for learning[edit | edit source]

Unite the enthusiastic programmers, and work on games, in which players can learn language and other subjects.

Prerequisites: programming knowledge, dictionaries.

[REALIZED: https://playinfluent.com/]

Humorous listening material[edit | edit source]

Humorous listening materials in many languages. It helps to enjoy learning languages.

Prerequisites: materials, readers.

Parsed Tatoeba[edit | edit source]

There should be a database about part of speech. Then build the mechanism for the computer to learn how to parse the sentences.

Prerequisites: programming knowledge.

Vocabulary matrix[edit | edit source]

Just give an example:

Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Mandarin Chinese Japanese Japanese Japanese Japanese Multiple languages
现代汉语单字字频: Character frequency list of Modern Chinese CC-CEDICT CC-CEDICT CC-CEDICT CC-CEDICT 萌典 萌典 萌典 萌典 Jisho.org Jisho.org Jisho.org Jisho.org 倉頡之友。馬來西亞
character frequency pronunciation definition HSK level classifier part of speech pronunciation definition radical part of speech pronunciation definition JLPT level Cangjie 5 code
1 de

of / ~'s (possessive particle) / (used after an attribute) / (used to form a nominal expression) / (used at the end of a declarative sentence for emphasis)

see 的士

really and truly

aim / clear

1

ㄉㄧˋ dì

ㄉㄜ˙ de

ㄉㄧˊ dí

箭靶的中心

1.結構助詞:➊ ​ 置於形容詞後。如:「美麗的風景」、「聰明的小孩」。➋ ​ 置於名詞或代名詞後,表示所屬、所有的關係。如:「我的書」、「太陽的光」。➌ ​置於修飾片語或子句後。如:「他寄來的信,我昨天收到。」、「那賣花的人沒零錢找。」➍ ​ 置於副詞後。同「地」。如:「慢慢的走」、「高高的飛」。 2.句尾助詞:置於句尾,表示肯定或加強的語氣。

確、真、實在。

確實的、可靠的。

Na-adjective, Suffix

Noun

てき

まと

bull's eye, mark, target, object, adjective ending N1 hapi

As you can see, 的 is a polyphone and a polyseme. There should be links between corresponding items: the part of speech, the pronunciation and the definition of the same source, but here on a wikitable it is not possible.

The list of databases is Language/Multiple-languages/Culture/Licensed-Free-Databases

The databases have different licenses, some have restrictions on redistribution. So, if someone wants to create a project about this, the only way is to let users download these databases and import them to the program (including updating) manually.

If you want to learn the vocabulary, you can get some, and memorise more proficiently. For example, you can select all adjectives in the same or in several HSK levels, turn them into Anki, and memorise them all in one time.

This can be the fundamental for building Vocabulary division and can be used to serve as an offline pop-up dictionary for it.

Prerequisites: programming knowledge.

Note for listening[edit | edit source]

Open the file in the audio player, when you don't understand, just press the “m” key, then the program will mark it with the audio still going on. Repeat the selections later.

Of course, there is a lag between your starting to not understand and pressing the button, starting to understand and pressing the button. You can set an advance value for repetition.

Prerequisites: programming knowledge.

[REALIZED: https://manual.audacityteam.org/man/edit_menu_labels.html]

Bidirectional textbook[edit | edit source]

A book with two languages, in which each language is taught with the other.

But I wonder how to keep them in the same pace after teaching pronouns.

This would suit language exchange participators.

Prerequisites: extensive knowledge of a language, language teaching knowledge.

Live quiz[edit | edit source]

A teacher is the streamer. There is a program implemented for answering questions. The teacher gives a question, then watchers (including robots) submit answers. The program counts the most popular answers. The teacher reviews them then.

Prerequisites: programming knowledge.

Database of collocation[edit | edit source]

An open database covers the main types of collocations. This is less noticed but still important.

Prerequisites: programming knowledge.

Volume standardisation[edit | edit source]

On volunteer-run sites, you can hear different people's pronunciations, the volume of which may be high and low. You can often hear recordings with too-high or too-low volumes. Thus, a program can be created for standardising the volume of multiple sound files.

Prerequisites: programming knowledge.

Lemma trainer[edit | edit source]

There are programs for training irregular verbs, grammatical genders, etc. such as Quizlet. There are still a lot of things can be done:

https://en.wiktionary.org/wiki/Category:Lemmas_subcategories_by_language

Someone can do something on this: all of this kind of training can be integrated in one program, based on Wiktionary.

Prerequisites: programming knowledge.

Rotely[edit | edit source]

Finally, it is disclosed.

An open-source program for displaying information from a TSV or CSV file to a window periodically, with spaced repetition.

Its use is not limited to language learning.

Prerequisites: programming knowledge.

Site of mistakes[edit | edit source]

List all kinds of mistakes language learners have made. Also, let users take a test about the type of mistake is made.

Prerequisites: programming knowledge.

Flashcards according to personal forgetting curve[edit | edit source]

Usually flashcard programs do spaced repetition according to a set period (1 min, 5 min, etc.), but what if memorsation can be customised for each individual with the help of artificial intelligence?

It will require the computer to analyse your accuracy and speed of reaction for each card.

Prerequisites: programming knowledge.

Parallel question database[edit | edit source]

Questions for language learning in multiple languages. Languages sharing the same lemma subcategory can share these same questions.

Reading comprehension is a question type that is still valid after translation for every language. When a reading comprehension is translated, it can be reproduced into multiple choice and cloze. If TTS is available, it may become a listening section.

There is a way: find licensed documents (under the Creative Commons, the GNU Free Documentation License or something alike) and translate them under identical licenses.

To motivate people to translate voluntarily, the process of translation can be practised by language learners, then other people including native speakers correct the translations. The translated versions will be published after proper corrections.

Prerequisites: programming knowledge, licensed materials, translators.

Rule-based quizzes[edit | edit source]

A quiz program with customised rules inside. For example: speaking (SR required), listening (TTS required) and writing numbers, inflections, negations, active-passive transforms, and etc..

Prerequisites: programming knowledge.

Visual example sentences[edit | edit source]

On this site:

https://www.dartmouth.edu/~deutsch/Grammatik/Grammatik.html

Example sentences are given in photos. This can be applied on any languages.

Bad examples can be also displayed with reasons.

Prerequisites: programming knowledge.

Conditioned reflex trainer[edit | edit source]

People take long time before having a good command of a new grammar category. A program can be developed to train this intensively.

Prerequisites: programming knowledge.

Grammar sandbox[edit | edit source]

Choose some words to make a sentence up, then see the correct grammar.

Prerequisites: grammar knowledge, programming knowledge.

Minecraft world[edit | edit source]

Create something in Minecraft with descriptions in multiple languages.

Text vocabulary evaluator[edit | edit source]

Evaluate the vocabulary of a text by word frequency, or input a website and wished vocabulary structure and fetch corresponding pages, or even read your known words from the central vocabulary bank (described below) and specify how many new words you want to learn and fetch pages catering for that.

This helps to instantly find out if the text is suitable for the user's vocabulary level. It can be even developed as a browser add-on. Also, a database about evaluated texts can be built.

Prerequisites: programming knowledge.

Central vocabulary bank[edit | edit source]

There are many different programs for learning vocabulary, but when you start using one of them, you have to make some effort to tell the program which part of the vocabulary you've already known. When you make progress in one program, you have to do the same in the others. If there is a central bank (merely TSV/CSV files) to tell every program about your vocabulary, it will save a lot of time.

Each word has two attributes: familiarity and completeness, which can be rated by the user. Familiarity means how easy you can recall it; completeness means how much portion of its meanings do you know. Each attributes can have values 0~3.

Prerequisites: consent from different developers, inflection rules, dictionaries for languages that don't use space as word separator, lists of separable verbs, lists of phrases, script converters.

Wikipedia title fetch[edit | edit source]

Create lists of vocabulary from titles of entries in different language versions in Wikipedia.

This is easily achievable: https://stackoverflow.com/questions/4420584/how-to-get-wikipedia-page-in-multi-languages

Wikidata pages are actually more convenient to use. But the problem is that they don't show titles with more than one script for each site.

Prerequisites: programming knowledge.

THIS IS GOING ON.

Reading speed trainer[edit | edit source]

This idea was actually formed long before but I accidentally missed it.

A program for practising reading speed. Words will appear on random places on the window and you need to read them as fast as possible, provided that you can understand them.

Prerequisites: programming knowledge.

Checklist of knowledge[edit | edit source]

List of what to learn. Clear and tidy. Great for grammar.

Prerequisites: collaboration.

Vocabulary expansion with typing[edit | edit source]

A program for fetching articles from websites and typing. Hotkeys can be used to add new words to learn (maybe into the central vocabulary bank).

It combines two practices: typing and reading, thus saves time.

Prerequisites: programming knowledge.

Word/phrase detector[edit | edit source]

Lots of languages have space as their word separator. Exceptions are:

  • No spaces between words (Chinese, Japanese, Thai)
  • A space is often placed inside a word, between syllables (Vietnamese)
  • Some words are separable (German, Dutch, Afrikaans, Hungarian, Chinese)
  • A phrase has more than one word inside it. These words produce new meanings when placed together.

Prerequisites: word lists, grammar knowledge, programming enowledge

Pictograph fomulation test[edit | edit source]

This is not for learning, but for studying.

Let people with different native languages formulate an article with emoji, pictographs and other symbols. See how similar or how different their works are.

Database integrator[edit | edit source]

There are a lot of dictionaries free for personal use. They are in different formats. A program can be developed to integrate them, so that they can be displayed in just one dictionary program.

Prerequisites: programming knowledge.

Synonym trainer[edit | edit source]

Program based on thesauri, for practicing specific words.

I haven't dived into thesauri. They look beautiful. I can't tell what else I can come up with.

Prerequisites: programming knowledge.

Recursive consulter[edit | edit source]

A program for consulting a dictionary and get the result. If you have a word list, you can fetch the result for each item on the list . The result will be stored in a TSV/CSV spreadsheet.

The interval between two queries should not be short, otherwise the website's servers may be overloaded or take action to block this bot.

Distributing large portion of dictionary protected by copyright can be illegal, so there's probably no legal room for this sort of collaboration. Collaboration can however happen on dictionaries with licenses.

Prerequisites: programming knowledge.

Popular culture database[edit | edit source]

Like Know Your Meme and Urban Dictionary, there should be a database for every culture. It may be lists of words at first.

Keyboard layouts for scripts[edit | edit source]

There are already ergonomic keyboard layouts with only one language taken into consideration. Though they may support multiple languages, they usually only focus on one language.

In my perspective, there can be a set of keyboards for a script, each language has a suitable variant, differences between variants are little.

A section of website on ergonomic keyboard layouts:

http://xahlee.info/kbd/keyboard_layout_keybinding.html

An introduction to designing keyboard layout:

https://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=KeybrdDesign

An example on designing Jawi keyboard:

http://www.pertanika.upm.edu.my/Pertanika%20PAPERS/JST%20Vol.%2018%20(2)%20Jul.%202010/10%20Pg%20271-292.pdf.

Prerequisites: mathematical knowledge, letter frequency list, bigram frequency list.

Vague memorisation[edit | edit source]

Learn new words without knowing their meanings. This can be done by practising typing according to a frequency list.

This is experimental. Now I am trying this to learn Korean.

Vocabulary heat map[edit | edit source]

A Program for inspecting your vocabulary: 100 words are presented in a 10 * 10 grid on a webpage; 100 links to 100 such webpages are presented in a 10 * 10 grid on another webpage; a few links to the last-mentioned webpages are presented on another webpage (totaling a few ten thousand words to be inspected, sufficient for native speaker's level). You can evaluate your familiarity with each word and visualise this by heat map of the grids.

“Central vocabulary bank” can store the evaluated familiarity with words.

Prerequisites: programming knowledge, frequency list

Associate words with similar sounds[edit | edit source]

A program: it shows a word, then you type all words with one or two different syllables from the original one. The range of words can be defined according to frequency lists. This will help you on listening: you can keep in mind how you may mishear.

Prerequisites: programming knowledge, frequency list, word list with phonetic notes

Geography all-in-one trainer[edit | edit source]

Make use of GIS data, create a program for practising geography: countries, administrative divisions, highest mountains, longest rivers, etc.

Data sources: https://gisgeography.com/best-free-gis-data-sources-raster-vector/

Vincent[edit | edit source]

Learn like a child[edit | edit source]

A child is able to learn a language only with the help of his parents.

Of course the child is learning very fast compared to an adult, but adults can use the same process.

Create a software using the same process with technologies like Artificial Intelligence...

The idea is to learn naturally without too many efforts:

  • Listen and repeat first
  • No grammar
  • Learn whole sentences, not single words...
  • Learn only what is useful to you every day
  • As soon as you reach a basic level, talk to people


gozde-agar[edit | edit source]

Everything about the Word Contest !

Would you like to attend a contest which is in Polyglot.com?

If yes lets create one together..

In my opinion we have to find sensible answers some questions. For example;

1- Why people want to join of the word contest? First of all people want to learn new words and want to speak with more words. Also the contests are always fun. We need to make learning new words fun.

2- How do we keep people's attention high? We can give points to the participants who were able to success to stay in the top 10 list for a week (its just an example, please share your advice).

3- Which language cards will the contest contain? We need to make sure it works for a while. Everyone should be able to quickly help to build its infrastructure. That's why we can choose English first. If everything is fine we can add other languages in it or can create new contests which named French Word Contest, Spanish Word Contest...

4- How do we find lots of words to create a contest? We can create a team for this project or we can want to help from the all members.

Each small duello can take place between two or three people. The person with the highest score is placed higher on the overall ranking.

Please ask your questions. You can even provide suggestions for your questions :D

5-What will be the interface of the contest page? The interface design work is as in the picture(draft1).

Draft 1.

nmesomtoChukwu[edit | edit source]

Hi Vincent.

For your question on how to improve the site:

It would be great if there were customised pages for each of the major languages of the world; Mandarin, English, Hindustan, Russian, Arabic, Spanish, Bengali, Malay, Portuguese, and French

https://www.fluentin3months.com/most-spoken-languages/


These customised pages would be designed based on the language and culture of those people for example:

• Spanish would have red as a theme color (red is a very special color to the Spanish) and the page would have little pictures depicting landmarks, cultural clothing, hats and other important aspects of the people or designed with names of all countries that speak the language like below.

• There would be a container displaying fun facts about the Spanish every day and another for a Spanish word for the day like a flashcard. There would be specialised chat rooms and a button where members can click to begin (and name) a chat room for specific subjects like a chat for festivals in Mexico. This would give opportunity for the member who opened the chatroom to decide whether the chatroom is open to everyone else (and therefore can be joined by everyone who wants to) or open on invitation (where other members would be invited by typing their account names or other members can click a button to join which would send a joining message to whoever opened the chat to either accept or refuse). If a member is not accepted or invited into a chatroom she/he can read the chat but not to be able to make a comment. However, this rule would not apply to the admins of the site who are given the power to read and comment on all chatrooms (to ensure compliance to the communities guidelines).


• The same would be done for the other major languages in each their specialised pages. For French the pictures could be perhaps the tower Eiffel, champs Elysee and other things like those. For Mandarin perhaps a picture of the great Wall of China, a cultural hat and cloth and other things. All the specialised pages would have their words of the day, facts of the day and chatrooms.

• For each specialised page, the design of the specialised page would be constant for all chatrooms, and so on.


Here's a prototype: https://docs.google.com/document/d/1IovWe4PhVOY1pNyaZM1pQQc-zDs9xz1Mnv-dBJ0BD5M/edit?usp=drivesdk&ouid=114748498551826027630 Idea language.jpg

Contributors

GrimPixel, Vincent, Gozde-agar, 112.36.194.132 and Safasamimiat


Create a new Lesson