Find your next open-source project

Search by name, purpose, or keyword — or paste a GitHub repository URL
All skills

digital-health-clinical-asr-setup

Verify your NVIDIA_API_KEY works by round-tripping a clinical sentence through Magpie TTS and Parakeet ASR.

  • Part of NVIDIA/skills
  • 1.9k installs
  • Machine Learning · Speech models

When to use it

You have an NVIDIA_API_KEY and a fresh Python environment but haven't confirmed the hosted speech stack actually responds.

For example

You say

Set up the Clinical ASR Flywheel for me — I have my NVIDIA_API_KEY exported and want to confirm the hosted TTS and ASR stack works before I start curating terms.

You get
  • Length-only key check
  • Installed Python dependencies
  • TTS-to-ASR smoke-test transcript

How it works

  • Shows a data-disclosure table naming NVIDIA NVCF and Merriam-Webster as the two external endpoints before any text or audio is sent.
  • The smoke test must actually run, not be deferred, since it gates whether later stages can reach the hosted stack.
  • Merriam-Webster lookup is optional; without it, Stage 2 falls back to Magpie G2P with weaker coverage on rare clinical terms.

Before installing

An NVIDIA_API_KEY from build.nvidia.com is required for the hosted Magpie TTS and Parakeet/Nemotron ASR calls; Python 3.10+ is also needed.