Inspiration
Hearing about the struggles of studying medicine at university.
What it does
Python application returns definitions of technical terms from lectures in real time, allowing students to quickly glance down to their screen for an instant medical dictionary definition. Could be used in other scenarios more widely within the medical sector, or with panopto recordings.
How we built it
- Microsoft Azure speech-to-text API
- Merriam-Webster Medical Dictionary API
- Cross-checking transcribed text for medical terms and returning definitions in real time
Challenges we ran into
- Getting live time updates smoothly and with minimal delay
- Using the dictionary API, and eliminating all edge cases (such as dictionary returning definition of closest word)
- Eliminating common words, which are most frequently used in non-medical sense
Accomplishments that we're proud of
- Genuinely helpful tool
- Sleek GUI
- LIVE definitions which work almost seamlessly
What we learned
- How fiddly tkinter is
- Python isn't ideal for this sort of application
What's next for Mediction
- Expand to other subjects! Like maths, which would require LaTeX support for equations
- Easily amendable filters for different words (maybe ability to star words you don't know and create a personal glossary)
- Optimising AI to learn from each voice
Built With
- azure
- merriam-webster-dictionary
- python
- speech-to-text
Log in or sign up for Devpost to join the conversation.