Our main concern was giving the ability to follow international news, and any other videos, to people with visual disabilities. As we progressed, we've noticed that this could be used in fields like Education, International Affairs, Entertainment, and Business.

What it does

Converting a mp3, or mp4 file's audio (dubbing) to desired language and embedding the translated audio in to the video in a higher pitch volume

How we built it

We've built it by using Watson's Speak to Text, Text to Speak API, Google Translate API, SSML

Challenges we ran into

Setting the timestamps of each dialog, limits of the Watson's Bluemix API, new speaking speed of speaker

Accomplishments that we're proud of

Dubbing a video as almost identical to it's original version

What we learned

Before we have started the application we thought the only challenge would be on using APIs, but as we progressed we have seen that was only the seen part of the iceberg. We faced countless issues about timestamp's of each dialog and we, once again, understood the importance of the planning phase of a project

What's next

Optimizing the timestamps, and rates which determines the speed of the dub

Share this project: