crafter-station

crafter-station / trx

Public

Agent-first CLI for audio/video transcription via Whisper

trx.crafter.run agent audio captions cli speech-to-text
40
2
100% credibility
Found Mar 31, 2026 at 40 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A command-line tool that transcribes audio and video from local files or online URLs into text, subtitles, and JSON using local AI models.

How It Works

1
🔍 Discover trx

You hear about a simple tool that turns videos or audio clips from anywhere into easy-to-read text transcripts.

2
⚙️ Set it up

You run a one-time setup that downloads what it needs and gets your computer ready to transcribe.

3
🎤 Feed it media

You give it a link to a video from YouTube or social media, or pick a file from your device.

4
🔄 It works its magic

The tool grabs the media, cleans up the sound, and converts speech to text automatically.

📄 Get your transcript

You receive clean text, timed subtitles, and other formats ready to copy, share, or edit.

Sign up to see the full architecture

3 more

Sign Up Free

Star Growth

See how this repo grew from 40 to 40 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is trx?

trx is a TypeScript CLI for transcribing audio/video from local files or URLs like YouTube, Twitter, and Instagram, powered by local Whisper models via whisper-cli. It handles downloads with yt-dlp, audio cleanup via ffmpeg, and outputs clean text, SRT, or agent-ready JSON. Developers get a one-command pipeline for machine-readable transcripts, skipping cloud APIs and setup headaches.

Why is it gaining traction?

Agent-first design sets it apart: JSON schemas for introspection, dry-run validation, and token-saving field filters make it pipe perfectly into AI workflows. Unlike basic Whisper wrappers, it auto-detects output formats, supports 99 languages, and bundles skills for agent post-processing on errors like accents or repeats. Social media URL support and local processing hook devs tired of API limits.

Who should use this?

AI agent builders scripting transcription pipelines, podcasters generating show notes from recordings, or researchers analyzing conference talks. Perfect for TypeScript/Bun users in agent-first GitHub orgs handling audio/video, not trx coin miners or trx training enthusiasts.

Verdict

Promising for agent-first audio/video tasks, but 40 stars and 1.0% credibility score signal early maturity—docs are crisp, yet expect tweaks. Try trx init if local Whisper fits; skip for production scale.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.