DeepgramAI Transcribe blockifier

This project contains a Steamship Blockifier that transcribes and analyzes audio files via DeepgramAI.

Configuration

This plugin is configured using an API key that is for running the Deepgram AI model. Those keys are supplied via secrets.

Getting Started

Usage

To authenticate with Steamship, install the Steamship CLI with:

> npm install -g @steamship/cli

And then login with:

> ship login

from steamship import Steamship, File, MimeTypes
from pathlib import Path

PLUGIN_HANDLE = "deepgram-s2t-blockifier"


ship = Steamship()  # Without arguments, credentials in ~/.steamship.json will be used.
audio_path = Path("FILL_IN")
s2t_plugin_instance = ship.use_plugin(plugin_handle=PLUGIN_HANDLE)
file = File.create(ship, content=audio_path.open("b").read(), mime_type=MimeTypes.MP3)
blockify_response = file.blockify(plugin_instance=s2t_plugin_instance.handle)
blockify_response.wait(max_timeout_s=3600, retry_delay_s=1)

file = file.refresh()

for block in file.blocks:
    print(block.text)

Developing

Development instructions are located in DEVELOPING.md

Testing

Testing instructions are located in TESTING.md

Deploying

Deployment instructions are located in DEPLOYING.md