ChatEase

Inspiration

ChatEase was inspired by the need for a simple and effective way to practice speaking in a foreign language. Many language learners struggle with active listening, real-time speaking, and remembering new phrases during conversations. We wanted to build a tool that makes it easier to take notes, understand what others are saying, and support multi-language interaction—all in one place.

Since OpenAI Whisper is multilingual and I'm using the tiny version for performance reasons, the results aren't perfect—but it is certainly usable and reliable enough for practical use.

What it does

ChatEase is a desktop application that allows users to:

Take notes during conversations (Japanese notes are automatically converted to romaji)
Translate between two languages (using Helsinki-NLP for EN-JA, JA-EN and Google Translate for others)
Convert speech to text using Whisper, capturing what the other person says in real time

It is especially useful for language practice on platforms like Discord.

How we built it

We used the following tools and technologies to build ChatEase:

Python 3.12.3 – main programming language
PyQt5 – for the graphical user interface
OpenAI Whisper – for accurate speech recognition
deep-translator – to access Google Translate API
langdetect – to automatically detect language of the input
Cutlet – to convert Japanese text into romaji
soundcard, soundfile, audioop – for audio recording and processing
gTTS – to generate text-to-speech responses (optional)

Challenges we ran into

Ensuring smooth real-time audio streaming and transcription
Managing language detection and routing text to the correct translation pipeline
Designing a user-friendly interface with multiple active components
Handling Japanese text processing accurately and reliably

Accomplishments that we're proud of

Seamlessly integrated Whisper with a PyQt GUI
Built a robust note-taking and translation workflow
Implemented real-time romaji conversion for Japanese learners
Created a practical tool that is already being used in real-time language practice