swstack / dialbot-whisper-api

API for ingesting large audio datasets and converting to text

Dialbot Whisper API

API for ingesting large audio datasets and converting to text'

Setup

Pre-requisites:

• Python 3.10+
• Poetry (https://python-poetry.org/docs/#installation)

Install python dependencies:

poetry install --no-root

Replace the .env file with the correct values for your environment.

Run the tests

Run the tests:

poetry run python tests/run.py

Running the application

There are various ways to run the application, including using the flask development server, gunicorn, or docker.

• Run with the flask development server:

  poetry run flask --app dialbot/run.py run

• Run with gunicorn:

  poetry run gunicorn --bind localhost:5000 --access-logfile - dialbot.run:app

• Run with docker:

  First build docker image:

  poetry export -f requirements.txt --output requirements.txt &&
  docker build --build-arg OPENAI_API_KEY=$OPENAI_API_KEY -t dialbot:latest .

  Run docker image:

  docker run -p 5000:5000 dialbot:latest