All skills
digital-health-clinical-asr-build
Curate clinical terms, tag IPA pronunciations, and synthesize a NeMo manifest with audio.
- Part of NVIDIA/skills
- 1.9k installs
- Machine Learning · Speech models
When to use it
You have a clinical specialty and a list of terms that ASR mistranscribes, but no benchmark audio or manifest yet.
For example
You say
Use digital-health-clinical-asr-build to build an oncology benchmark for chemo drug names like pembrolizumab and cisplatin, with clean and noisy audio.
You get
- term_seed.csv
- manifest.jsonl
- audio wav clips
How it works
- Terms pass through a three-tier IPA pipeline: verified overrides, then Merriam-Webster, then Magpie G2P, each tagged in the manifest.
- A QA audition gate synthesizes one clip per term before the full voice, noise, and context product is generated.
Before installing
NVIDIA_API_KEY is required for hosted Magpie TTS, and Stage 1 (/digital-health-clinical-asr-setup) must be completed first.
Also in this pack
- accelerated-computing-cudf3.2kGuides GPU DataFrame work with cuDF, including pandas acceleration, joins, groupby, file I/O and multi-GPU jobs.
- cuopt-developer2.8kModifies, builds, tests, and debugs NVIDIA cuOpt source code and prepares contributions.
- cuopt-install2.6kInstalls cuOpt for Python, C, or server via pip, conda, or Docker and verifies it works.
- data-designer2.6kGenerates a synthetic dataset from your description of the data you need.