-
-
Vid2Text Logo
-
Landing Page
-
Application Page
Inspiration 💡
With the introduction of online-based learning, a lot of video tutorials are being created for students and learners to gain knowledge. As much as the idea is excellent, there was also a constraint whereby the tutorials may contain long hours of content and in some cases, it is inaccessible to users with disability. This is seen as a problem in the world today, that's why we built an innovative and creative solution Vid2Text to provide the solution to this. It is a web application that provides users with easy access to audio and video text transcription for all types of users. So either if the file is in an audio or video format it can always be converted to readable text.
🍁About
Vid2Text is a web app that allows users to upload audio and video files with ease, which then generates Modified and Customized Audio and Video Transcriptions. Some of the features it provides are:
Features
- Automatically transcribe audio and video files with high accuracy.
- Modified and Customized Audio and Video Transcriptions.
- Easy Keyword Search through Text
- Easy Keyword Search and Highlight through Text.
How we built it
We built our project using Django, a Python web framework that uses the MVC architecture to develop full-stack web applications. When the user uploads the video they want to transcribe, we created a script that uploads the video onto the Django model database and after that, the video gets uploaded to the AssemblyAI server. The response of that part is the upload_url. Finally, we send a post request, with the video transcript ID and get the video transcript text as the response. We utilized the AssemblyAI to match and search the transcript text for keywords. We also created an accessible and good user experience for the client-side.
Challenges we ran into
In course of the hackathon, we faced some issues with limited time and integration of the AssemblyAI to determine the video duration of the uploaded videos dynamically. Initially, we were confused about how to do that, but we finally figured it out.
Accomplishments that we're proud of
Finally, after long hours of work, we were able to build and deploy the full web application. The team was able to put in extra effort towards making it work.
What we learned
This hackathon gave us the opportunity to learn how to use a Django project together with utilizing the Django and AssemblyAI API and also we were able to work together as a team despite the fact we were from different timezones.
What's next for Vid2Text ⏭
For our next steps: We plan to include more features like multiple language transcription, and export text files as pdf. Also, improve the user experience and make it more accessible.
Log in or sign up for Devpost to join the conversation.