Speech to Text Web App

Speech to Text Web App...

About

This is a web app that allows users to record audio and convert it to text. The app is built with Python / Django and React using several libraries.

Run locally with Docker

You might use sudo before commands if you are not in the docker group.

1. Clone the repository and cd into directory

$ git clone https://github.com/thascius-rumpelschnick/speech-text-web-app.git
$ cd speech-text-web-app

2. Add .env.dev file

Look at the .env.example file and create a .env.dev file with the same variables. Or see here for a starter and search for heading Dev Environment.

3. Build frontend assets

$ docker-compose run --rm node sh -c "npm i && npm run build-dev"

4. Load Vosk model

$ docker-compose run --rm python sh -c "python manage.py load_vosk_model -m vosk-model-small-de-0.15"

5. Start database and make migrations

$ docker-compose up -d postgres
$ docker-compose run --rm python sh -c "python manage.py migrate"

6. Start application as whole

$ docker-compose up -d postgres python

7. Stop application

$ docker-compose down

You can now access the web application at http://localhost:8000.

After you are done with steps 1 to 5 for the first time, you just need to run steps 6 & 7 from then on.

Screenshots

Author

Florian Zapf

Docs

1. Development
2. ToDo's