Ctrlk
Page cover

Translation Station

In this Scratch project, you will create an exciting Translation Station - your very own custom tool to input phrases and translate them, both in writing and speaking.

Your objective is to code and learn about artificial intelligence (AI) by creating a dynamic game. In this safe and creative method to use AI, you can input a word or sentence in a given language and have it translated to the language of your choice.

Press the Green Flag to begin!

🎒 About this lesson plan

Students build a simple translation tool in Scratch where they can type any English word or phrase, select a target language from a dropdown menu, and watch AI instantly translate and pronounce it.

Students can test their AI translator with increasingly challenging words, explore languages outside of language study classes, and get feedback from multilingual classmates on pronunciation accuracy. Overall, students will discover where AI translation succeeds or stumbles. They'll uncover which languages work better than others, why some cultural concepts don't translate directly, and how their simple tool connects them to millions of speakers worldwide.

1

Setup

⏳5 minutes

Students will create their own translation tool in Scratch to help study for French class.

  1. Open Scratch 3.0 and create new project

  2. Add AI translation blocks from extensions

  3. Create script

2

Core Activity - Vocabulary Victory

⏳25 minutes

Students build and test their translator, then save it as a study tool they can use anytime.

About Vocabulary Victory

Students pair up to practice vocabulary for their language classes. One student picks a word from their study list and selects the target language, while their partner (without looking at the screen) tries to pronounce that word in the new language. Then they press space to hear the AI's pronunciation and see if they were close. Partners switch roles and keep score of correct guesses.

  1. Build Your Study Tool (15 min)

  2. Code: when space pressed → translate English input to French (or other language being studied) → speak result

  3. Test with actual French vocabulary: "butterfly," "house," "friendship"

  4. Partner up: one picks English word, other guesses French pronunciation, press space to check

  5. Save & Practice (10 min)

  6. Save project with a clear name like "French Quiz Helper"

  7. Practice with real vocabulary from upcoming French test

  8. Keep score of correct pronunciations and notice what's tricky for AI

3

Extension - Language Detective

⏳15 minutes

Students explore other languages and discover how well the tool works within their home languages.

About Language Detective

Students test the AI by translating a word, then translating it back to English. Does "butterfly" → Spanish → English still say "butterfly"? Students become detectives finding where AI translation "breaks" and discuss why.

Multilingual students in the classroom can become experts who can verify translations and spot what the AI gets right or wrong. They might catch that while "butterfly" translates correctly, "social butterfly" doesn't make sense in their language. These student experts might introduce words from their native languages that don't exist in English, challenging classmates to understand concepts that the AI struggles with.

  1. Free Exploration (10 min)

  2. Test any languages students are curious about or speak at home as multilingual learners

  3. Try family words, slang, or phrases from different cultures

  4. Group Discovery Sharing (5 min)

  5. Groups share one interesting discovery

  6. What worked well? What seemed off? Any funny mistranslations?

4

Reflection & AI discussion

⏳15 minutes

Students reflect on their discoveries and discuss how AI learns languages.

  1. What Did We Learn? (5 min) - Students share translation accuracy discoveries

  2. AI Training Talk (10 min) - Teacher-led discussion:

  3. Why wasn't pronunciation always perfect?

  4. Which languages worked better and why?

  5. What does this tell us about AI training data and whose languages get priority?

Student Check-In Tool

✅ Demystifies AI - With Translation Station, students discover how AI translation works by testing it, finding its limitations, and understanding what training data means for language representation.

Students can track their progress with the check-in table below, simply circle or mark an X in the corresponding box!

Still Working

Getting There

I Got It!

I can build a working translator

I can test my translator

I can describe one strength and one limitation of AI translation

🤝Curriculum Connections

🎯Math Grades 3-4: C3.1 solve problems and create computational representations of mathematical situations by writing and executing code, including code that involves sequential, concurrent, and repeating events

🎯Math Grades 5-6: C3.1 solve problems and create computational representations of mathematical situations by writing and executing efficient code, including code that involves conditional statements and other control structures

🔗 Links to Scratch Projects

Starter code

Completed Scratch Project

Extended version

Code-related Objectives:

  • Introduce learners to conditional statements with event blocks

  • Introduce learners to the use of artificial intelligence tools in code.

  • Represent mathematical and linguistic situations computationally

  • Test and debug code

⚙️Tools

  • Computer

  • Internet Connection

  • Browser (Chrome, Firefox, Safari, Edge)

  • Scratch website.

  • Optional: Speakers or headphones for audio pronounciation

🍃Sustainable Development Goals

This Scratch activity relates to some of the UN's Sustainable Development Goals, mainly:

Visit to learn more: https://sdgs.un.org/goals

🍎 Teacher's Corner

  • Scratch also works on tablets and other mobile devices but you may need to make some changes to the code.

  • Prior to the workshop, it is recommended to:

Before we dive into Let's Play, let's go over some Fun Facts!

NextFun Facts

Last updated