skills.rest
reasonlayer
reasonlayerOFFICIAL
Explore all skills by this author →
GitHub

use-local-whisper

Switch WhatsApp voice transcription to a local whisper.cpp pipeline on Apple Silicon.

Updated Mar 5, 2026
Search Tags:#whatsapp#ffmpeg#on-device#whisper.cpp#apple-silicon#voice-transcription
One-click install
npx skills add https://github.com/reasonlayer/rl-agent --skill use-local-whisper-reasonlayer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: use-local-whisper
Source: https://github.com/reasonlayer/rl-agent/tree/main/.claude/skills/use-local-whisper
Command: npx skills add https://github.com/reasonlayer/rl-agent --skill use-local-whisper-reasonlayer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Switches voice transcription from external APIs to a local whisper.cpp-based pipeline, enabling on-device processing and offline operation.

Core Features & Use Cases

  • Local transcription using whisper.cpp on Apple Silicon devices, reducing API key reliance and network usage.
  • Audio format conversion with ffmpeg to 16 kHz mono WAV for reliable transcription.
  • Seamless integration with the existing voice-transcription skill to handle media downloads from messages.

Quick Start

Invoke this skill to transcribe a WhatsApp voice message locally using whisper.cpp on-device.

Frequently Asked Questions about use-local-whisper

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

FAQPage Schema
How do I transcribe WhatsApp voice messages locally on my Mac?▼

Local WhatsApp voice transcription uses a whisper.cpp pipeline running on-device with Apple Silicon. Audio is converted with ffmpeg to 16 kHz mono WAV for reliable transcription, requiring a GGML model and the voice-transcription skill to handle media downloads.

Can I use whisper.cpp for offline voice transcription without an API key?▼

Yes, whisper.cpp enables offline voice transcription without an API key by processing audio entirely on-device. It replaces external transcription APIs with a local pipeline, requiring only a GGML model and ffmpeg for audio format conversion.

What do I need to run on-device whisper transcription for WhatsApp audio?▼

On-device whisper transcription requires whisper.cpp, a GGML model, and ffmpeg installed on an Apple Silicon architecture. You must also apply the voice-transcription skill first to handle media downloads from WhatsApp messages.

Does local whisper transcription work on platforms other than Apple Silicon?▼

Local whisper transcription is specifically designed for Apple Silicon architectures. Other messaging channels and hardware platforms require their own audio handling implementations to work with the whisper.cpp pipeline.

Why does ffmpeg need to convert audio to 16 kHz mono WAV for whisper.cpp?▼

Ffmpeg converts audio to 16 kHz mono WAV because whisper.cpp requires this specific format for reliable transcription. Standardizing the audio format ensures the on-device model processes the voice messages consistently.

Related Skills

Explore More Software Engineering Skills→
alanmunalanmun

add-voice-transcription

Transcribe WhatsApp voice messages using OpenAI's Whisper API.

Community
Intermediate
smilebank7smilebank7
2

Translate

Translate transcribed lecture text into a target language while preserving timestamps.

Community
Intermediate
clsandovalclsandoval

use-local-whisper

Transcribe WhatsApp voice messages locally using whisper.cpp on Apple Silicon.

Community
Advanced
okbest2024okbest2024

openai-whisper

Transcribe audio locally using the Whisper CLI without external APIs.

Community
Basic