Gettr Transcribe (MLX Whisper) for Openclaw

A high-performance local transcription tool for GETTR content optimized for Apple Silicon using MLX Whisper.

kevin37li
v1.0.1
Feb 23, 2026
0
1.3k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install gettr-transcribe

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install gettr-transcribe using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Gettr Transcribe (MLX Whisper)?

This skill provides a streamlined workflow for downloading and transcribing audio from GETTR posts and streaming pages. By utilizing MLX Whisper, it leverages the power of Apple Silicon to perform high-quality, local speech-to-text conversion without the need for cloud-based services. As part of the Openclaw Skills ecosystem, it handles the complexities of media extraction, audio normalization, and timestamped transcription generation.

The tool is specifically designed for developers and researchers who need to process GETTR media into VTT format. It handles both direct audio links and HLS streams, ensuring that the transcription pipeline is robust enough for various media types found on the platform. This makes it an essential component for any automated media analysis workflow using Openclaw Skills.

Gettr Transcribe (MLX Whisper) Use Cases

  • Archiving GETTR live streams into searchable text documents for research.
  • Generating VTT subtitle files for GETTR video posts to improve accessibility.
  • Local transcription of social media content to maintain data privacy.
  • Converting non-English GETTR broadcasts into text using multi-language Openclaw Skills support.

How Gettr Transcribe (MLX Whisper) Works

  1. The user provides a GETTR URL, and the skill identifies the unique content slug.
  2. Browser automation is used to navigate the page and extract the underlying media URL (either .m4a or .m3u8).
  3. The media is downloaded and converted into a standardized 16kHz mono WAV format using ffmpeg.
  4. MLX Whisper processes the audio file locally to produce a transcript with precise timestamps in VTT format.

Gettr Transcribe (MLX Whisper) Setup

To get started with this skill, ensure you have the necessary dependencies installed on your Apple Silicon machine:

# Install MLX Whisper for local transcription
pip install mlx-whisper

# Install ffmpeg for media processing
brew install ffmpeg

# Run the pipeline with a media URL and slug
bash scripts/run_pipeline.sh "<AUDIO_OR_VIDEO_URL>" "<SLUG>"

Gettr Transcribe (MLX Whisper) Data Schema & Taxonomy

The skill organizes transcription outputs into a structured directory based on the GETTR content slug. This taxonomy ensures easy integration with other Openclaw Skills for downstream processing like summarization.

File Path Description
./out/gettr-transcribe/<slug>/audio.wav The normalized 16kHz mono audio file extracted from the source.
./out/gettr-transcribe/<slug>/audio.vtt The final transcript file including WebVTT timestamps for media alignment.

Gettr Transcribe (MLX Whisper) Advanced Features

  • Support for explicit language tagging (e.g., --language zh) to improve accuracy on non-English content.
  • Flexible model selection allowing users to switch between tiny, base, medium, or large-v3 Whisper models.
  • Automated handling of expired signed URLs via browser automation re-runs.
  • Optimization for Apple Silicon (M-series chips) for significantly faster local inference compared to standard CPU processing.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*