Skip to content

Commit

Permalink
Merge pull request #40 from dahabibi/patch-1
Browse files Browse the repository at this point in the history
Update README.md
  • Loading branch information
MAlshaik authored Apr 12, 2023
2 parents 36d3ef1 + 73e2883 commit 7d6d927
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,15 +16,15 @@ The user interface of the application is built using HTML, CSS, and JavaScript.

## Transcription

The transcription subsystem is powered by Open AI's Whisper model. When a user uploads a video, the system automatically transcribes the audio and displays it on the website. The interactive transcript makes it easy for users to navigate through their video and find specific parts of the audio.
We initially wanted to use Amazon Transcribe for the transcription, but Amazon charged for usage of the service. The transcription subsystem is powered by Open AI's Whisper model. Specifically, a fork of Whisper that tracks the time of each word in the video. When a user uploads a video, the system automatically transcribes the audio and displays it on the website. The interactive transcript makes it easy for users to navigate through their video and find specific parts of the audio.

## Video Timeline

Originally, we planned to use a JavaScript library called amalia.js for the video timeline. However, due to the library's newness, we ultimately decided to build our own timeline using CSS and JavaScript. The timeline allows users to edit their videos quickly and easily, without having to worry about the technical details of the editing process.

## Video Editing

The video editing process is handled by the backend using Python's MoviePy library. Users can delete specific words in the transcript to remove portions of the video they don't want to include. This process is simple and intuitive, allowing users to make edits quickly and easily.
The video editing process is handled by the backend using Python's MoviePy library. Users can delete specific words in the transcript to remove portions of the video they don't want to include. The process recombines the video parts that weren't deleting, creating a new video. This process is simple and intuitive, allowing users to make edits quickly and easily.

## Technologies Used

Expand Down

0 comments on commit 7d6d927

Please sign in to comment.