Inspiration
The problem came into my mind when I was looking for a research topic and the results I got was in a language I could not understand, well I don't know if that video was actually useful for me or not but the concern was
What it does
So it basically identifies the language of the source video and converts it to any language you prefer keeping into consideration the gender, age, and character of the person speaking
How we built it
Django was used to build the backend and Bootstrap was used for the frontend. The reason for using Django is that it supports Python as a backend and can use external libraries.
Challenges we ran into
The main issue was detecting the gender of the input voice and based on gender detect the voice as well detect multiple characters. Another main challenge was detecting the breaks in the input voice and based on which align the translated script.
Accomplishments that we're proud of
The moment we saw a working prototype we were relieved, the main accomplishment was working on the problem of the break the character took while delivering dialogues.
What we learned
Using NLP to recognize the words that were not clear in translation and making them perfect using lemmatizing and using deciotnary
What's next for Honor
To align the face movement with translation.
Log in or sign up for Devpost to join the conversation.