Spoken Tutorial

Motivation

Built for the screening phase of the FOSSEE program under Python.

Technologies/Frameworks Used

• ffmpeg-python
• Django
• pyttsx3

Why function based views instead of class based views?
Personal preference. I find function based views more clear and verbose in what they do as compared to class based (magic) views.

Why pyttsx3?
Other options were gTTS which did not have male/female voice options, and Google Cloud TTS / IBM Watson, neither of which are easily accessible (require banking information). So this module became the go to choice.

Wireframe

Instructions

Pre-requisites

Download ffmpeg

Windows

Mac

Linux - just run sudo apt-get install ffmpeg or equivalent for your distro.

ffmpeg should be in your $PATH, video tutorial for windows.

Install requirements

pip install -r requirements.txt

Run migrate command

python manage.py migrate

Running

Run using

python manage.py runserver

Open in browser

Visit localhost.

Screenshots

How to use?

• Default page is the merge portal, upload valid video and audio files.
• It should give a Preview button after clicking merge, if something went wrong either ffmpeg is not in path or uploaded files are invalid.
• The TTS page accepts text file with readable text.

Example contents: "The quick brown fox jumped over a branch."

• If the text file is valid a download button would be visible.
• Video Demo