The goal of the speech recognition flashcard software from FlipStudy was to make it easier for student to study and hone their pronunciation skills. Traditional flashcard apps frequently depend on users rating their own pronunciation, which can be subjective and may not be an accurate representation of their level of proficiency. The FlipStudy app gives users a more objective and interactive learning experience by integrating voice recognition technology. By offering an enjoyable, interesting, and personalized learning experience, we hope to assist users in achieving their language learning objectives.
What it does
Displays text that the user is learning and asked them to speak it to which it will listen and tell if you you are pronouncing it correctly.
How we built it
Programmed in Python and using WordPress for the website framework.
Challenges we ran into
We did not know how to setup and use speech recognition so we found a speak recognition library for Python which we read documentation to use to build a speech interpreter. We also ran into trouble integrating WordPress with Python as our backend was mainly PHP.
Accomplishments that we're proud of
This is our first time integrating PHP and Python to work together.
What we learned
Working with Speech to Text and Text to Speech over API and with Python. Calling Python from PHP and returning the results for PHP to work with further. Building audio recorders in HTML5.
What's next for FlipStudy
We hope to introduce languages beyond English. Speech to Text processing with supervised based machine learning. With a human reviewing the sound bytes, then they can provide the feedback the AI needs to adapt to various speech patterns.
Log in or sign up for Devpost to join the conversation.