skills.rest
TobiasWooldridge
TobiasWooldridgeCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

wavecap-llm

Configure LLM-based post-processing to correct Whisper audio transcriptions via YAML.

Updated Aug 27, 2026
Search Tags:#ai#llm#transcription#whisper#apple silicon#correction
One-click install
npx skills add https://github.com/TobiasWooldridge/WaveCap --skill wavecap-llm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: wavecap-llm
Source: https://github.com/TobiasWooldridge/WaveCap/tree/main/.claude/skills/wavecap-llm
Command: npx skills add https://github.com/TobiasWooldridge/WaveCap --skill wavecap-llm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses inaccuracies and jargon in automated transcriptions by leveraging local Large Language Models (LLMs) for intelligent correction, ensuring higher fidelity and domain-specific accuracy.

Core Features & Use Cases

  • LLM-based Correction: Automatically corrects errors in Whisper transcriptions using configurable local LLMs.
  • Model & Prompt Tuning: Allows users to select different LLM models, adjust generation parameters (temperature, max tokens), and define domain-specific terms to preserve jargon.
  • Use Case: A medical professional needs highly accurate transcriptions of patient consultations. This Skill can be configured with a suitable LLM and domain terms like medical abbreviations to ensure the final transcriptions are precise and professional.

Quick Start

Configure the wavecap-llm skill to enable LLM correction using the 'llama-3.2-3b' model and set the temperature to 0.1.

Frequently Asked Questions about wavecap-llm

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

FAQPage Schema
How do I correct Whisper transcription errors using a local LLM?▼

You can correct Whisper transcription errors using a local LLM by configuring LLM-based post-processing. This Skill enables intelligent correction of automated transcriptions to ensure higher fidelity and domain-specific accuracy.

Can I preserve domain-specific jargon when applying LLM correction to audio transcriptions?▼

Yes, you can preserve domain-specific jargon when applying LLM correction to audio transcriptions. The Skill allows you to define domain-specific terms, ensuring specialized vocabulary like medical abbreviations remains intact during correction.

Does LLM-based transcription correction work on Apple Silicon?▼

Yes, LLM-based transcription correction works on Apple Silicon. The Skill specifically targets Apple Silicon hardware and requires the mlx-lm library to run local model inference for transcription enhancement.

How do I configure a specific LLM model like llama-3.2-3b for transcription post-processing?▼

You configure a specific LLM model like llama-3.2-3b for transcription post-processing via a YAML file. This configuration file manages model selection, generation parameters, and domain term preservation settings.

What are the limitations of using local LLMs for Whisper output correction?▼

Limitations of using local LLMs for Whisper output correction include the requirement for mlx-lm for local model inference and Apple Silicon hardware. Generation parameters like temperature and max tokens must also be carefully tuned to avoid altering original audio meaning.

Related Skills

Explore More Software Engineering Skills→
J4JoshuaJ4Joshua

mlx

Optimize local AI inference on Apple Silicon using MLX acceleration.

Community
Advanced
devMoezdevMoez
1

whisper

Converts spoken audio into text with optional English translation and timestamps.

Community
Intermediate
aresbitaresbit
46

openai-whisper

Transcribe audio locally using the Whisper CLI with configurable models and output formats.

Community
Advanced
malue-aimalue-ai
36

mlx-whisper

Transcribe audio files locally on Apple Silicon using MLX Whisper.

Official
Intermediate