whatsapp-voice-transcribe

Transcribe WhatsApp voice notes and apply mining-site term corrections.

1|Updated Jul 30, 2025
One-click install
npx skills add https://github.com/karstegg/MarthaVault --skill whatsapp-voice-transcribe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: whatsapp-voice-transcribe
Source: https://github.com/karstegg/MarthaVault/tree/main/.claude/skills/whatsapp-transcribe
Command: npx skills add https://github.com/karstegg/MarthaVault --skill whatsapp-voice-transcribe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the tedious process of transcribing WhatsApp voice notes, especially for technical or industry-specific content like mining site names, which often get mis-transcribed by generic models. It saves time and ensures accuracy by applying specialized corrections.

Core Features & Use Cases

  • Automated Transcription: Downloads and transcribes WhatsApp voice notes using advanced Whisper MCP technology.
  • Industry-Specific Corrections: Applies intelligent corrections for common mining site mis-transcriptions (e.g., "Trinity" to "Nchwaning 3", "Hutt" to "Haut").
  • Vault Integration: Securely moves transcribed audio files to a designated vault media folder for organized storage and easy retrieval.
  • Use Case: A user receives several voice notes from colleagues discussing critical site operations. Instead of manually listening and typing, they can ask Claude to "transcribe my latest voice note from Jacques Breet" and instantly receive a corrected, accurate text summary, ready for reporting or action.

Quick Start

Transcribe my latest WhatsApp voice note from Jacques Breet and apply mining site corrections.

Frequently Asked Questions about whatsapp-voice-transcribe

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I transcribe WhatsApp voice notes to text?

Transcribe WhatsApp voice notes by requesting Claude to download messages from a specific contact, process them with Whisper MCP, and output accurate text. The Skill automates downloading, transcription, and optional file vault storage in a single workflow.

Can I correct mining site names during transcription?

Yes. The Skill applies a mining-site corrections dictionary post-transcription to fix common mis-transcriptions like "Trinity" to "Nchwaning 3" and "Hutt" to "Haut", ensuring industry-specific accuracy without manual editing.

What do I need to use WhatsApp voice transcription?

You need Filesystem MCP to move files, Whisper MCP with gpt-4o-mini-transcribe for transcription, and access to WhatsApp messages. The Skill handles coordination; ensure your environment has these dependencies configured.

How does the Skill organize transcribed audio files?

After transcription, the Skill securely moves audio files to a designated vault media folder, organizing them for easy retrieval and maintaining a clean working directory while preserving the original content.

Can transcription work with voice notes from multiple contacts?

Yes. Specify the contact name when requesting transcription—for example, "transcribe my latest voice note from Jacques Breet"—and the Skill downloads and processes messages from that specific contact only.

What happens if mining site names aren't in the corrections dictionary?

The Skill applies corrections only for terms in its mining-site dictionary. Unmatched terms pass through as transcribed; review and manually correct specialized site names not covered by the default corrections.