Speech Collect

A speech collection tool for machine learning datasets

First time setup

Create a virtual environment and activate it. Now from the root project directory run ./scripts/bootstrap. This will install pip-tools and sync any dependencies for the first time.

Dependency management

Dependencies are managed via [pip-tools].

Adding a dependency

To add a dependency, edit requirements.in (or dev-requirements.in for dev dependencies) and add your dependency then run pip-compile requirements.in.

Syncing dependencies

Run pip-sync requirements.txt dev_requirements.txt.

Testing

Run the tests by invoking py.test in the project root. Make sure you run any pending migrations beforehand.

Local Development Environment

This project makes use of kind and Tilt as a local development environment. To ensure you have the system dependencies installed run:

check-system-deps

Missing dependencies can be installed from their official sources (alternatively you can use homebrew for most):

• kubctl | brew install kubectl
• helm docs | brew install helm
• Docker running as non-root
  • Docker Install
  • Docker Post Install Steps
• kind installation | brew install kind
• ctlptl | brew install tilt-dev/tap/ctlptl
• python3.11 deadsnakes | brew install python3.11
• tilt bash download | brew install tilt

To start the local development environment and navigate to the :

make start-local