summarize-audio

Summarize Audio (Async Chaining)

Audio transcription support not yet considered fully-production ready on Steamship. We are working hard on productionizing support for audio transcription at scale, but there may be some existing issues that you encounter as you try this out.

This is an example of deploying an app LangChain for summarization on Steamship. It provides an integration with one of Steamship's audio transcription plugins and a small introduction to the Task system.

Try it out

Or, install dependencies:

pip install steamship-langchain
pip install termcolor

And run the client:

python client/client.py

Deploy your own

Switch to the server/ directory and run deploy.

cd server
ship deploy

The deployment script will walk you through setting up a package name that you can use for your own instance, if desired. This will enable you to modify the example to meet your needs (or just to have fun experimenting with).

After deployment, switch back to the parent directory ($ cd ..) to run the client, etc. You'll need to update the package_handle in the client to match your new deployment.

api.py

api.py is required by Steamship for packages being deployed. You may add additional source files as desired, but you MUST always have api.py.

A note on dependencies

Steamship relies on requirements.txt as part of the packaging and deploy. If you add new dependencies to your server code, please ensure they are reflected in requirements.txt.