skills.rest
Alamator
AlamatorCOMMUNITY
Explore all skills by this author →
GitHub

voice-refine

Convert verbose voice dictation into concise, structured Claude prompts with context, objective, constraints, and outputs.

Updated Apr 12, 2026
Search Tags:#voice#cleanup#transcription#compression#speech-to-text#prompt-engineering
One-click install
npx skills add https://github.com/Alamator/code-ultimate --skill voice-refine-alamator
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: voice-refine
Source: https://github.com/Alamator/code-ultimate/tree/main/examples/skills/voice-refine
Command: npx skills add https://github.com/Alamator/code-ultimate --skill voice-refine-alamator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Voice-refine turns long, unstructured speech-to-text transcripts and voice memos into concise, token-efficient prompts that preserve intent while eliminating filler, repetitions, and tangents.

Core Features & Use Cases

  • Deduplication & Cleanup: Removes filler words, hedging, and repeated phrases to clarify intent.
  • Requirement Extraction & Structuring: Identifies core requirements, constraints, and expected outputs and organizes them into Context, Objective, Constraints, and Output sections.
  • Compression & Language Handling: Compresses content toward ~30% of original length while retaining intent and supports preserving or switching output language; ideal for cleaning dictation, meeting notes, bug reports, and feature requests.

Quick Start

Use the voice-refine skill to convert a long speech-to-text transcript into a concise, structured Claude prompt including context, one-sentence objective, constraints, and expected output.

Frequently Asked Questions about voice-refine

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

FAQPage Schema
How do I turn speech-to-text transcripts into structured Claude prompts?▼

Transform speech-to-text transcripts into structured Claude prompts by applying voice-refine to deduplicate content, extract core requirements, and compress the dictation into context, objective, constraints, and output sections.

What is the best way to clean up voice memos for prompt engineering?▼

The best way to clean up voice memos for prompt engineering is to remove filler words, hedging, and tangents while organizing the remaining intent into concise, sectioned prompts that preserve the original requirements.

Can I use voice-refine on short dictation outputs under 150 words?▼

Voice-refine targets dictation outputs longer than approximately 150 words containing filler words, repetitions, or tangents, so shorter transcriptions may not benefit from the deduplication and compression logic.

How do I compress voice transcription while preserving the original intent?▼

Compress voice transcription while preserving intent by extracting core requirements, structuring the content into dedicated sections, and reducing the overall length to about 30% of the original size.

Does voice-refine support switching the output language of my voice notes?▼

Voice-refine supports preserving or switching the output language of your voice notes, allowing you to clean dictation and meeting notes while generating the final structured prompt in your preferred language.

Related Skills

Explore More Software Engineering Skills→
alexica00alexica00

voice-refine

Convert verbose voice input into structured Claude prompts with Context, Objective, Constraints, and Output.

Community
Intermediate
DethonDethon
3

prompt-creator

Enhance rough prompt ideas into detailed, codebase-aware prompts.

Community
Intermediate
t-miura-024t-miura-024

mt-format-voice-input

Format raw voice transcription text with corrected punctuation and paragraph structure.

Community
Intermediate
samhvw8samhvw8
14

prompt-enhancer

Transform verbose prompts into concise, XML-structured versions for multiple AI models.

Community
Basic