Skip to content

jevy/transcribe-and-burn-subtitles

 
 

Repository files navigation

AWS Transcribe and FFMPEG burning of subtitles

This tool will take a video:

  1. Transcribe it using AWS
  2. Allow you to proof read and edit the result
  3. Burn the subtitles into a video using FFMPEG.

This is useful for sites like Facebook or LinkedIn where the audio doesn't auto play but you want to grab the viewer's attention.

This is a heavy modification/simplification on this lovely repository: https://github.com/aws-samples/aws-transcribe-captioning-tools

Get the transcription

  1. Create an S3 bucket for your input videos and the output videos
  2. Put the videos you want to translate into the vids_to_process dir.
  3. Run transcribeVideos.py

Edit it

  1. Edit the generated subtitles in the subtitles dir

Build the video

  1. Run pip install -r requirements.txt
  2. Run the burnSubtitles.py script. It requires you have docker up and running

About

Convert AWS Transcribe output into multiple caption formats.

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 97.3%
  • Batchfile 2.6%
  • Dockerfile 0.1%