Django Speech-to-Text app

A simple Django app that allows users to upload and transcribe audio files in a Django app.

Demo:

A step-by-step tutorial can be found in the corresponding blog post.

Get Started

Step 1: Install dependencies:

pip install Django assemblyai python-dotenv

Step 2: Rename /stt_project/.env.template to /stt_project/.env and enter your AssemblyAI API key. You can get a free API key here.

ASSEMBLYAI_API_KEY=your_key_here

Step 3: Run the server

cd stt_project
python manage.py runserver

App architecture

These were the steps to create this demo from scratch. We created one Django project (called stt_project)and one Django app (called transcriptions):

django-admin startproject stt_project
cd stt_project
python manage.py startapp transcriptions

The following files were modified:

• The view that handles transcriptions is implemented in /transcriptions/views.py
• The template to upload and submit audio files, and to show the transcribed text, is implemented in /transcriptions/templates/index.html
• The URLs are defined in /stt_project/urls.py and /transcriptions/urls.py
• Environment variables are loaded with load_dotenv() in /stt_project/settings.py

Improvements

For more advanced implementations with non-blocking options, read this section in the tutorial blog post.