Google has launched new language tools providing free personalised French lessons powered by AI.
The online platform, known as Little Language Lessons, claims to offer “a collection of bite-sized learning experiments, built with Gemini [Google’s AI assistant].”
It features three tools that allow users to give specific prompts, in turn generating relevant example phrases and vocabulary in a target language.
Exercises are currently offered in 15 languages such as French, Italian, German, Arabic and Japanese. Regional variations of certain languages are also included, for example Spanish (Latin America) and Spanish (Spain).
“These experiments are not about replacing traditional study, but about complementing it: helping people build habits, stay engaged, and weave learning into their everyday lives,” said Aaron Wade, Creative Technologist at Google.
Read also: Five social media accounts to follow to improve your French
The Connexion tried out each of these tools to see how French learners could benefit from them.
How to use the tools?
The platform is free and easy to access - all you need is an internet connection and a Google account.
After signing in, a pop-up appears on-screen stating: “Little Language Lessons is an early-stage experiment. Although its intent is to help language learners, this tool uses generative AI—so its outputs may not always be accurate or complete. We recommend cross-referencing with reliable sources.”
Once accepting this message, you can choose between three tools: Tiny Lesson, Slang Hang, or Word Cam.
Tiny Lesson
The first experiment generates relevant vocabulary, phrases, and grammar tips for any situation, for example, booking a hotel room, finding a lost passport or taking a taxi.
You start by selecting your target language - in this case, French (France), although French (Canada) is also an option.
A purpose or theme can then be typed in the following box. For best results, focus on a specific situation rather than a general context. We tried the prompt: ordering a baguette.
Gemini then generates a language guide, split into vocabulary, phrases and general tips.
The vocabulary list features six words, which can also be expanded to offer a further six examples. Key terms included: la baguette (the baguette), le pain (the bread), la boulangerie (the bakery).
A particularly useful tool is the ability to hear the pronunciation of each example by clicking on the dedicated blue speaker button. The French voice is generally very accurate and can allow learners to practise recreating the appropriate sounds.
Read also: Five podcasts to help your French listening skills
The example phrases mainly stemmed from the previously listed vocabulary, but also anticipated other common sentences used in a bakery:
Je voudrais une baguette, s'il vous plaît. - I would like a baguette, please
Combien coûte une baguette? - How much does a baguette cost?
Je prends aussi un pain au chocolat. - I'll also take a pain au chocolat.
Finally, the grammar tips were helpful and accurate, highlighting the importance of politeness in French conversation, shown by using the conditional mood as well as saying merci and s'il vous plaît.
Helpful explanations also indicated how to talk about ‘uncountable’ quantities of food, for example, using du pain instead of le pain.
Read also: Letters: French is a hard language - even for native speakers
We did however notice some errors and confusions. For example, requesting ‘une baguette entière’ (a whole baguette) does not exist in French and would likely be met with a raised eyebrow.
Similarly, the phrase ‘avec ceci, ce sera tout’ combined both a question typically asked by the baker (avec ceci?) with the customer’s response (ce sera tout).
Numbers also appeared in their numerical form, rather than being spelled out, which relies on learners having prior knowledge of the French translations.
Overall, the results were basic and failed to capture the nuances of ordering in a boulangerie with no mention of baguette types or baking techniques. The tool acts more as an entry into a specific language context, intended for casual use.
For full baguette-ordering tips, visit our dedicated language article.
Slang Hang
The second experiment focuses on idioms, expressions and regional terms that could be heard in casual French conversations.
This tool does not allow you to personalise a situation, but rather generates a random context based on the selected language: “Two strangers, Sophie and Antoine, find themselves unexpectedly sharing a park bench during a sudden downpour in Paris. Sophie, a graphic designer, is sketching in her notebook, while Antoine, a botanist, is carrying a delicate orchid. They strike up a conversation as they wait for the rain to subside.”
The conversation is broken into speech bubbles, with some lines featuring a colloquial word or phrase which is underlined. By hovering your mouse over this line, a box will pop-up to explain the meaning behind it.
Despite the tool being called ‘Slang Hang’, the generated conversation in our experience did not feature any slang references - the result was actually very formal. It rather focused on French turns of phrase and expressions such as:
La pluie s'est mise à tomber d'un coup. - It suddenly started raining.
Si je ne m'abuse. - If I am not mistaken.
C'est gentil à vous. - That is kind of you.
We tried generating more conversations, some of which did feature informal slang terms in varying weird and wacky situations - which did not always feel life-like.
Read also: DeepL, Reverso: Which French-English translation app do you use?
This tool would not be useful for beginners as it requires quite an advanced level of French - an extra button must be pressed to produce English translations. It could however allow learners to read and hear how a conversation could naturally unfold between two French speakers.
Word Cam
The third and final experiment allows users to open their camera and capture their surroundings to learn new words in a given context.
We tried taking a photo of The Connexion’s most recent print edition. The AI was able to identify le journal (the newspaper), la photo (the photograph on the front page), and la plante (the plant on the desk beside the paper).
You can then click on one of the headings to generate more related vocab examples.
After clicking on le journal, adjectives such as quotidien (daily), informatif (informative) and local (local) were generated. Accompanying sentences showed how to use these words to talk about newspapers in French.
Word Cam could be handy in situations where you need to name a specific object, however we found that better language results and examples were generated using the Tiny Lesson tool.
To find out more about how these tools were created, visit Google’s explanatory blog.
The rise of AI in language learning
This Google Labs development could be useful for learners seeking quick phrases when visiting a new country, and makes an engaging and interactive change to standard textbook-driven lessons.
The tools will require further development if Google wants to offer comprehensive and accurate resources to its users.
The launch of this experimental platform comes as AI continues to integrate into the development of language-learning platforms to create a personalised experience.
Read also: Improve your French: Our pick of the best language learning apps for 2025
Popular language-learning app, Duolingo, also recently announced that it is evolving to become an ‘AI-first’ model. This move aims to increase access to new markets and languages, hoping the adjustment will allow them to reach one billion users around the world.
Duolingo already cut 10% of its contracted translators in January 2024 as it began to transition to AI-powered services, reports The Telegraph, and the growth of AI suggests this trend could continue.
In a post on Linkedin, Luis von Ahn, CEO of Duolingo, wrote: “Developing our first 100 courses took about 12 years, and now, in about a year, we are able to create and launch nearly 150 new courses. This is a great example of how generative AI can directly benefit our learners.”
Commenters expressed their disapproval of the company’s new direction, citing incoherence as well as a disregard for ethics and culture. Many stated they would be leaving the app altogether.
