Seethroughme

SeeThroughMe provides useful tools for travelers who need help identifying and translating signs or objects in foreign countries. The app takes your image submission, converts it to descriptions, translates them to different languages, and produces pronunciations for you to listen to. The application can be used in both web and mobile platforms, making it flexible for travelers on the go.

Tech Stack: React.js, Node.js, Express.js Technologies: IBM Watson Visual Recognition, IBM Watson Text to Speech, Google Tesseract OCR, Google Translation

• Augmented machine learning capacity by complementing the ‘IBM Watson Visual Recognition’ and ‘Google Translate’ technologies with ‘IBM Watson Speech to Text’ and ‘Google’s Tesseract Optical Character Recognition engine’ to enable pronunciations of translated signs or images of objects • Enriched overall UX with intuitive country-specific buttons for language selection and an auto-scroll carousel for a cumulative collection display • Devised Mocha/Chai tests in adherence to the TDD paradigm to enforce application stability and performance