Problem statement 4. Transcribing text to speech sounded good, since we have done machine learning together before.

What it does

Allows upload of wav (audio) and mp4 (video) files. Automatically transcribes the files into text, allowing search by content.

How we built it

We started with a failed attempt to use DJango. When we tried PHP, we got much farther than DJango in very little time. So we switched to PHP, and had some successes near the end.

Challenges we ran into

Django is outrageously hard to use. For example, we spent a number of hours trying to get an image to display on a DJango-hosted webpage. We determined that it involved modifying several files. And even after we tried to modify those files, the image didn't come through.

Accomplishments that we're proud of

Getting something to work.

What we learned

Tricks for using PHP. Not to use DJango. That I can stay up until 4am if I need to.

What's next for VeTranscriber

Filtering and sorting of search results.

Share this project: