Inspiration
For Sam and Vedarsh's parents, English was not their first language. When they were learning English, they encountered many challenges when learning how to pronounce English words correctly. So, we decided to create a web app that helps English learners improve their pronunciation.
What it does
English Pronunciation Helper records the user speaking a preset passage and provides feedback on how clear their pronunciation was. After recording the user in-browser, English Pronunciation Helper uses the Assembly AI speech-to-text API to get the words that were pronounced clearly. It then compares those words with the words from the original passage and shows the user which words they pronounced clearly and which words they did not. The user can choose the difficulty of the passage they want to speak.
How we built it
Assembly AI, React, JavaScript, Web Speech API, Google Firebase
Challenges we ran into
Finding the tools to construct our project, getting our API to work properly, and integrating Assembly AI
Accomplishments that we're proud of
Recording audio in-browser and the algorithm we wrote to determine the differences between what the user spoke and the original passage.
What we learned
How to use the Assembly AI API and integrate Firebase into a React app.
What's next for English Pronunciation Helper
We would like to improve the UI, expand the library of passages, and provide a quantitative score of the user's pronunciation.
Built With
- assemblyai
- firebase
- javascript
- react
- webspeechapi
Log in or sign up for Devpost to join the conversation.