Member-only story
Building an AI-Powered Language Learning App: Learning From Two AI Chatting
A step-by-step tutorial on creating a dual-chatbot language learning app with Langchain, OpenAI, gTTS, and Streamlit
When I first began learning a new language, I like to buy those “conversational dialogues” books. I find those books very useful as they help me understand how the language worked — not just the grammar and vocabulary, but also how people really used it in day-to-day life.
Now with the rise of large language models (LLMs), a thought occurred to me: could I replicate these language-learning books in a more interactive, dynamic, and scalable format? Could I utilize LLM to create a tool that generates fresh, on-demand conversations for language learners?
This thought inspired the project I would like to share with you today — an AI-powered language learning app, where learners can observe and learn from two AI chatbots engaged in either a user-defined conversation or a debate.
Regarding the employed tech stack, I have used Langchain, OpenAI API, gTTS, and Streamlit to create the application where users can define the roles, scenarios, or debate topics, and let the AI generate the content.
