skills.rest
RavitejaKarra24
RavitejaKarra24COMMUNITY
Explore all skills by this author →
𝕏@RavitejaKarra24GitHub

transcribe

Transcribes audio files to timestamped text locally on Apple Silicon macOS.

Updated Mar 12, 2026
Search Tags:#offline#transcription#macos#audio#ffmpeg#speech-to-text#apple-silicon
One-click install
npx skills add https://github.com/RavitejaKarra24/dotfiles --skill transcribe-ravitejakarra24
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: transcribe
Source: https://github.com/RavitejaKarra24/dotfiles/tree/main/pi/.pi/agent/skills/pi-skills/transcribe
Command: npx skills add https://github.com/RavitejaKarra24/dotfiles --skill transcribe-ravitejakarra24

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ffmpeg, curl, tar.

What problem does it solve? Converting audio recordings into text usually requires uploading files to a cloud service. This Skill performs speech-to-text transcription entirely on-device using the parakeet-cpp-transcribe binary, keeping audio data local on Apple Silicon Macs. ## Core Features & Use Cases - Local Speech-to-Text: Runs the parakeet-cpp-transcribe engine with an automatically downloaded GGUF model, no cloud API required. - Timestamped Output: Produces plain-text transcript lines chunked into 15-second segments like [00:00-00:15] text. - Format Handling: Accepts WAV directly and converts other audio formats via ffmpeg to 16kHz mono before transcription. - Use Case: Record a meeting or voice memo, then run the transcribe script on the file to get a timestamped text transcript without sending audio to any external service. ## Quick Start Transcribe the audio file meeting-recording.m4a into timestamped text using the transcribe script.

Frequently Asked Questions about transcribe

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

FAQPage Schema
How do I transcribe an audio file to text locally on macOS?▼

Run transcribe.sh with the audio file path as its single argument. The script downloads the parakeet-cpp-transcribe binary on first run and prints timestamped transcript lines to stdout in 15-second chunks.

What audio formats does local speech transcription support?▼

WAV files are transcribed directly. Other formats such as m4a or mp3 are converted automatically to 16kHz mono WAV using ffmpeg, which must be installed via brew install ffmpeg.

Does this transcription tool work on Intel Macs or Linux?▼

No. The script explicitly checks the platform and exits with an error unless it is running on Darwin arm64, meaning Apple Silicon macOS only.

Why is ffmpeg required for transcription of some audio files?▼

The transcription engine only accepts WAV input. For non-WAV files, ffmpeg converts the audio to 16kHz mono WAV first; without ffmpeg installed, the script exits with an error for those formats.

How do I hide diagnostic logs during audio transcription?▼

Model and GGML diagnostic logs are written to stderr while the transcript goes to stdout. Redirect stderr with 2>/dev/null to see only the timestamped transcript lines.

Related Skills

Explore More Software Engineering Skills→
CerisonAutomationCerisonAutomation

transcribe

Transcribe audio and video files into text using a configured speech-to-text provider.

Community
Intermediate
JYeswakJYeswak

granola-extract

Extract decisions and action items from meeting notes with source-line citations.

Community
Intermediate
metric-space-aimetric-space-ai
2

transcribe

Transcribe audio to text with optional speaker diarization and known-speaker hints.

Official
Intermediate
spikelabspikelab

transcribe

Transcribe audio files to text using mlx-whisper on Apple Silicon macOS.

Community
Intermediate