
Digital health clinical asr build
Builds a clinical term benchmark with synthesized audio and a NeMo manifest for ASR tests
- Scripts
- None
- Platforms
- Claude Code, Codex, Cursor
- GitHub stars
- 4K
- Last commit
- 3 days ago
- License
- Apache-2.0
- Listing updated
- 1 day ago
Similar skills
About Digital health clinical asr build
Digital health clinical asr build is an open-source agent skill for health and science. It is available on Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf and Goose.
This stage of the Clinical ASR Flywheel helps build an evaluation set for medical speech recognition. The agent interviews the user about their specialty and the terms that cause transcription errors, then records them in a term list tagged with categories such as drug, procedure, anatomy, condition, lab and role.
The workflow generates short clinical sentences for each term in several context types, such as dictation, handoff, chart note and history. Pronunciations are resolved in tiers: a team-maintained override file first, then a Merriam-Webster respelling converted to IPA, and finally the TTS engine's own grapheme-to-phoneme model. Candidate IPA strings are checked against the synthesis engine's phoneme set, and a single-clip audition step is required before the full set of audio is generated.
The output is a folder with WAV clips at clean and noisy signal levels, a NeMo-format manifest with clinical metadata, and an appendable pronunciation override file. Running it requires an NVIDIA API key for hosted Magpie TTS, an optional Merriam-Webster dictionary key, and a completed setup stage. Only synthetic, non-patient text should be sent to these external services.
When your agent uses it
Stage 2 of the Clinical ASR Flywheel. Use when curating clinical terms, tagging IPA, and synthesizing a NeMo manifest. NOT for scoring (use /digital-health-clinical-asr-eval).
From the skill's SKILL.md. Your agent reads this description to decide when to load the skill.
What it can access
From a static check of its 7 files (59 KB) on GitHub on 17 November 2026. Nothing was run. Read the files before you install a skill: it can tell your agent to run anything its tools allow.
Check another skill before you install it.
Install
Copy the skill folder to ~/.claude/skills/digital-health-clinical-asr-build to use it in all your projects:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/NVIDIA/skills.git "$tmp" &&
mkdir -p ~/.claude/skills &&
rm -rf ~/.claude/skills/digital-health-clinical-asr-build &&
cp -R "$tmp/skills/digital-health-clinical-asr-build" ~/.claude/skills/digital-health-clinical-asr-build &&
rm -rf "$tmp"For one project only, put the folder in .claude/skills/digital-health-clinical-asr-build inside it instead.
Copy the skill folder to ~/.agents/skills/digital-health-clinical-asr-build to use it in every repository:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/NVIDIA/skills.git "$tmp" &&
mkdir -p ~/.agents/skills &&
rm -rf ~/.agents/skills/digital-health-clinical-asr-build &&
cp -R "$tmp/skills/digital-health-clinical-asr-build" ~/.agents/skills/digital-health-clinical-asr-build &&
rm -rf "$tmp"For one repository only, put the folder in .agents/skills/digital-health-clinical-asr-build inside it instead.
Copy the skill folder to ~/.cursor/skills/digital-health-clinical-asr-build to use it in every project:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/NVIDIA/skills.git "$tmp" &&
mkdir -p ~/.cursor/skills &&
rm -rf ~/.cursor/skills/digital-health-clinical-asr-build &&
cp -R "$tmp/skills/digital-health-clinical-asr-build" ~/.cursor/skills/digital-health-clinical-asr-build &&
rm -rf "$tmp"For one project only, put the folder in .cursor/skills/digital-health-clinical-asr-build inside it instead.
Run in a terminal (add --scope workspace to install it for the current workspace only):
gemini skills install https://github.com/NVIDIA/skills.git --path skills/digital-health-clinical-asr-buildOr copy the folder to ~/.gemini/skills/digital-health-clinical-asr-build yourself.
For GitHub Copilot in VS Code, copy the skill folder to ~/.copilot/skills/digital-health-clinical-asr-build:
tmp=$(mktemp -d) &&
git clone --depth 1 https://github.com/NVIDIA/skills.git "$tmp" &&
mkdir -p ~/.copilot/skills &&
rm -rf ~/.copilot/skills/digital-health-clinical-asr-build &&
cp -R "$tmp/skills/digital-health-clinical-asr-build" ~/.copilot/skills/digital-health-clinical-asr-build &&
rm -rf "$tmp"For one project only, put the folder in .github/skills/digital-health-clinical-asr-build inside it instead.
Get the folder:
git clone --depth 1 https://github.com/NVIDIA/skills.gitThen copy skills/skills/digital-health-clinical-asr-build into your agent's skills folder as digital-health-clinical-asr-build. Agents that support SKILL.md read one folder per skill.
What it does
- Clinical term curation
- Guides a short specialty interview and records terms with a fixed set of entity categories.
- Tiered pronunciation lookup
- Resolves IPA from overrides, Merriam-Webster respellings or the TTS model's built-in G2P, in that order.
- Sentence generation by context
- Produces several sentence variants per term across dictation, handoff, chart note and history contexts.
- Audition gate before synthesis
- Synthesizes one clip per term for listening review before the full benchmark is generated.
- NeMo manifest output
- Writes a JSONL manifest with clinical fields and noise-level variants alongside the WAV audio.
- Appendable pronunciation overrides
- Keeps a verified IPA override file that carries across evaluation cycles.
Good for
- Build a benchmark of oncology drug names for speech recognition testing
- Generate eval audio for ICU handoff terminology
- Add orthopedic procedure terms to an existing cycle manifest
- Create a NeMo manifest from a list of cardiology medications
- Audition TTS pronunciations for hard-to-say drug names
Digital health clinical asr build FAQ
What does this skill need before it runs?
It needs the setup stage completed, an NVIDIA API key for hosted Magpie TTS, and optionally a Merriam-Webster dictionary API key for pronunciation lookup.
Does it send patient data to external services?
No. The skill tells users to send only non-PHI synthetic content, since term lists and generated sentences go to Merriam-Webster and NVIDIA's TTS service.
What files does it produce?
A cycle folder with audio WAV clips, a manifest.jsonl in NeMo format, the term seed CSV and a pronunciation override CSV.
What comes next after this stage?
The next stage is the digital-health-clinical-asr-eval skill, which transcribes the manifest and scores results, using KER as the headline metric.
How do I install Digital health clinical asr build in Claude Code?
Copy the skill folder from GitHub to ~/.claude/skills/digital-health-clinical-asr-build to use it in all your projects, or to .claude/skills/digital-health-clinical-asr-build in one project.
Does Digital health clinical asr build run scripts?
No. It is instructions only, with no scripts. This comes from a static check of the files on GitHub; nothing was run. Read the files before you install it.
Which AI agents can use Digital health clinical asr build?
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf and Goose. Each reads skills in the SKILL.md format from its own folder; the install steps on this page list them.
What are the best alternatives to Digital health clinical asr build?
The closest alternatives to Digital health clinical asr build are Digital health clinical asr eval, Speech, Digital health clinical asr finetune, Transcribe and Digital health clinical asr setup.
Where this comes from
- Description: from the skill's own SKILL.md.
More from NVIDIA/skills

Guides an agent in converting cuTile GPU kernels to Triton and debugging the result
- Open source
- 4K GitHub stars
- Developer tools

Adds autotuning with cached config selection to CuTile GPU kernels
- Open source
- 4K GitHub stars
- Developer tools

Guides an agent to write, debug and validate cuTile GPU kernels in Python
- Open source
- 4K GitHub stars
- Developer tools

Guides an agent through iterative profiling and tuning of cuTile GPU kernels in TileGym
- Open source
- 4K GitHub stars
- Developer tools

Guides an agent to swap TileGym cuTile kernels into Hugging Face transformers models
- Open source
- 4K GitHub stars
- Developer tools

Composes and deploys an NVIDIA VSS vision stack with captioning, detection, search
- Open source
- 4K GitHub stars
- Design and media




