Google is strategically advancing towards entering a new market, teaching languages to people. Duolingo has been the trustworthy and go-to application for people interested in learning a new language. However, Google has announced it’s launching a new AI-powered feature on its Google Translate platform designed to help people to learn and practice a new language. This decision is aimed at challenging Duolingo’s dominance in the market.
The new language practice feature is designed for both beginners starting to learn conversational skills and advanced speakers looking to brush up on their vocabulary. According to Google, to do so, it creates tailored listening and speaking practice sessions that adapt to a user’s skill level and unique learning goals. Moreover, the gamified approach used in the feature can help people learn around 40 languages.
How does the new feature work?
Users can set their fluency level—from Basic to Advanced—and choose goals such as everyday interactions or travel scenarios. The app then generates listening and speaking exercises tailored to that level. For example, if the prompt is “Ask about meal times,” users can either listen to simulated conversations and tap the words they understand, or practise speaking with hints when needed.
Progress is tracked daily, encouraging consistent practice. For now, the feature is available in beta to English speakers learning Spanish and French, and vice versa for Spanish, French, and Portuguese speakers learning English. It’s free during the testing phase.
How to access this feature?
Users interested in accessing the feature will select the “practice” option in the Google Translate app. From there, one can set skill level and goals.
AI integration in Live Translate?
Google’s Gemini AI model, which the company says improves translation accuracy, multimodal understanding, and text-to-speech output has been used in the new feature. Live Translate, which allows users to hold real-time, back-and-forth conversations with both audio and on-screen translations, will now be more accurate thanks to Gemini. Supporting over 70 languages—including Hindi, Tamil, Arabic, French, and Spanish—the mode uses a thread-like interface to intelligently detect pauses, accents, and intonations. Google says it has trained its models to filter out background noise, making it useful in crowded or busy environments.
Google’s Stance on the new feature
Google wrote in its blog post. “As we continue to push the boundaries of language processing and understanding, we are able to serve a wider range of languages and improve the quality and speed of translations. And with our Gemini models in Translate, we’ve been able to take huge strides in translation quality, multimodal translation, and text-to-speech (TTS) capabilities.”
