Design & media

plaud-transcripts-corrected-with-your-own-terminology-glossary

Try it

Get correct terminology in every Plaud transcript by maintaining your own glossary.

What it does

Applies a Google Sheets glossary of your terms to Plaud recordings before any summary or action items are generated. Each run reads new recordings, corrects them against your vocabulary and formatting rules (such as Swiss German spelling or thousands separators), then outputs a Google Doc with the corrected transcript, summary, corrections log, and any unknown terms flagged for glossary expansion.

When to use it

  • Lab meeting where cell line and reagent names keep coming back misspelled in transcripts
  • Legal or case recordings where matter numbers and client names are mangled by speech recognition
  • Technical team calls where part numbers and local spelling conventions are consistently misheard
  • Ongoing recording archive that needs consistent terminology across summaries and follow-ups

The skill document

Plaud Transcripts Corrected With Your Own Terminology Glossary

Freshness

Last updated: 2026-07-26.

If the current date is more than 7 days after the last updated date, reinstall this skill from skills.sh or ClawHub before relying on endpoints, schemas, setup steps, or examples.

What This Workflow Does

Fixes the words your transcription keeps getting wrong, by giving the pipeline your vocabulary instead of hoping a bigger model guesses right. Every speech model mangles terms it has never seen: cell line and reagent names, drug and device names, case and matter numbers, part numbers, local spelling and number conventions, team and client names. Swapping to a different model does not fix this, because none of them have your terms either. Keep a plain Google Sheet glossary of the terms that matter, each with the spelling you want and the misfires it usually comes back as ("HEK293" for "heck two ninety three"), plus any formatting rules you need enforced, such as Swiss German ss instead of eszett or 1'000.00 number style. On each run the workflow reads the glossary, pulls the transcript of every new Plaud recording, applies the corrections to the transcript first, and only then writes the summary and action items from the corrected text, so your terminology is right everywhere downstream instead of only in the raw transcript. The corrected transcript and summary land in a Google Doc, and every correction made is listed so you can see what changed and keep tuning the glossary. Unknown terms are flagged, never silently normalized into something that looks plausible.

Required Setup

  • AgentPMT overview: ../what-is-agentpmt.
  • Account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup.

Core AgentPMT setup skills:

  • What AgentPMT is: ../what-is-agentpmt
  • AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup

skills.sh install script:

npx skills add AgentPMT/agent-skills --skill what-is-agentpmt
npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup

Workflow Session Management

Call AgentPMT-Workflow-Skills with start_workflow before the first step and end_workflow after the final step.

{"action":"start_workflow","skill_id":"plaud-transcripts-corrected-with-your-own-terminology-glossary"}
{"action":"end_workflow","skill_id":"plaud-transcripts-corrected-with-your-own-terminology-glossary","rating":5,"comment":"completed"}

Workflow Process

  1. Get User Timezone and Date
  2. Ask for the Recording Window
    • Prompt: Establish the exact window of recordings this run covers, in the user's own timezone.
  3. Read Terminology Glossary
    • Tool product: Google Sheets.
    • Tool skill: ../google-sheets.
    • ClawHub page: https://clawhub.ai/agentpmt/google-sheets.
    • skills.sh install: npx skills add AgentPMT/agent-skills --skill google-sheets.
    • Marketplace: https://www.agentpmt.com/marketplace/google-sheets-api.
    • Tool instructions: Read the glossary tab: each row carries the correct term, the misheard variants it usually comes back as, and optional notes. Also read any formatting rules rows (spelling conventions, number formats). This sheet is the single source of the user's vocabulary.
  4. List Plaud Recordings
    • Tool product: Plaud.
    • Tool skill: ../plaud.
    • ClawHub page: https://clawhub.ai/agentpmt/plaud.
    • skills.sh install: npx skills add AgentPMT/agent-skills --skill plaud.
    • Marketplace: https://www.agentpmt.com/marketplace/plaud.
    • Tool instructions: Call list_files with the widened UTC date_from and date_to from the window step. Plaud's timestamps are inconsistent and must be handled deliberately: the name field carries the recording's LOCAL time (for example '2026-07-25 19:53:50') while start_at and created_at are UTC written as naive strings with no Z and no offset (the same recording reads '2026-07-25T23:53:50'). Treat start_at as UTC, convert it into the user's timezone, and only then decide whether the recording falls inside the requested local window. An evening local recording carries the next day's UTC date, so comparing the raw string silently misfiles it. Then skip any recording id already in the processed log.
  5. Each New Recording
    • Iterate over the configured collection, then continue through the connected workflow path.
  6. Fetch Transcript
  7. Summarize the Run
    • Prompt: Report which recordings were corrected and which terms the glossary is still missing.
  8. Apply Glossary to Transcript
    • Prompt: Correct the transcript against the user's glossary and formatting rules before any summary is produced.
  9. Summarize From Corrected Text
    • Prompt: Produce the summary and action items from the corrected transcript so the user's terminology is right downstream, not just in the raw text.
  10. Write Corrected Doc
  • Tool product: Google Docs Connector.
  • Tool skill: ../google-docs-connector.
  • ClawHub page: https://clawhub.ai/agentpmt/google-docs-connector.
  • skills.sh install: npx skills add AgentPMT/agent-skills --skill google-docs-connector.
  • Marketplace: https://www.agentpmt.com/marketplace/google-docs-connector.
  • Tool instructions: Create a document for this recording, headed with its local recorded date and time, containing the summary and action items, then the corrected transcript, then a corrections table listing each change and the glossary row behind it, and finally any unknown terms flagged for the user to add to the glossary.
  1. Log the Run
  • Tool product: Google Sheets.
  • Tool skill: ../google-sheets.
  • ClawHub page: https://clawhub.ai/agentpmt/google-sheets.
  • skills.sh install: npx skills add AgentPMT/agent-skills --skill google-sheets.
  • Marketplace: https://www.agentpmt.com/marketplace/google-sheets-api.
  • Tool instructions: Append the recording id, its local recorded time, the doc link, the number of corrections applied, and any flagged unknown terms to the processed log so the recording is never handled twice and the glossary can be tuned over time.

AgentPMT Reference

Questions people ask

What happens if a term isn't in the glossary?
The workflow flags it as unknown rather than silently guessing. You can then add it to the glossary for the next run.
Does this work with any speech-to-text provider?
The workflow fetches transcripts from Plaud and applies corrections on top of whatever Plaud's transcription produces.
How does the glossary handle formatting rules like number styles or local spelling?
The glossary sheet can include formatting rule rows alongside term pairs, covering conventions like 1'000.00 number format or Swiss German spelling.

Related skills

Make irreversible life decisions by projecting to 80 and naming which regret you'd rather live with.

by deciqai1 installs2 stars

Diagnose which mental domain is holding you back before choosing a cognitive intervention.

by deciqai1 installs3 stars

Structured framework to analyze your alternatives and acceptable deal range before you commit

by deciqai1 installs3 stars

Operate TaskTime Pro through MCP: manage tasks, track time, handle expenses, and prepare invoices from a paired browser session.

by tasktimepro1 installs1 stars

Detect when presentation language is steering your decision instead of the facts themselves.

by deciqai1 installs2 stars

Measure whether a transformation changed your growth engine or just added a one-time bump.

by deciqai1 installs2 stars

More from agentpmt

Browse all skills

Cloud-based Blender: render 3D models, generate turntable videos, and convert file formats without local installs.

by agentpmt3 installs

Extract text, structured entities, and metadata from any PDF, image, or scanned document.

by agentpmt3 installs

Generate install-ready Minecraft Bedrock add-ons, skin packs, Fabric and NeoForge mods with runtime verification and visual proof.

by agentpmt2 installs

Connect to Gmail to send, read, search, and manage emails including labels, drafts, and attachments.

by agentpmt1 installs