hoodini

hoodini / blitzai

Public

קול — Professional Transcription Studio. Hebrew-first, 4 engines, YouTube support, correction studio.

47
16
100% credibility
Found Apr 05, 2026 at 47 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Blitz AI is an open-source desktop app for transcribing audio and video from files, URLs, or folders into editable, timestamped text with Hebrew optimization and export options.

How It Works

1
🔍 Discover Blitz AI

You hear about this friendly tool that turns your audio recordings and videos into readable text, perfect for Hebrew talks or lessons.

2
🚀 Start the app

Follow easy pictures to get the transcription studio running on your computer in minutes.

3
Add your media
📎
Drag files

Drop audio clips or video files right onto the screen.

🔗
Paste link

Copy a video webpage address and it grabs the sound automatically.

📁
Pick folder

Choose a whole folder of podcasts or course videos.

4
⚙️ Set preferences

Choose your language like Hebrew and a speedy listener option.

5
✨ Watch it transcribe

Relax as the magic turns speech into timed text before your eyes.

6
🎬 Edit with video

Play the video synced to text, fix words easily, see who spoke what.

📤 Get your subtitles

Download perfect text files ready for videos, notes, or sharing.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 47 to 47 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 blitzai?

Blitzai is a self-hosted transcription studio that handles audio and video from local files, YouTube playlists, or folders, outputting timed subtitles in SRT, VTT, ASS, TXT, or JSON. Built with Python/FastAPI backend and Next.js frontend, it supports four engines—local Whisper, Groq, Gemini, ivrit-ai—for Hebrew-first accuracy with RTL UI and a synced video correction editor. Developers get a no-subscription alternative to Otter.ai or Descript, running entirely offline or with cheap cloud APIs.

Why is it gaining traction?

It stands out with Hebrew-optimized models like ivrit-ai, batch playlist processing via yt-dlp, and a pro studio for editing with word-level timestamps, confidence scores, and infinite undo. Unlike MacWhisper's basic playback or cloud tools' paywalls, blitzai mixes free local runs with 299x faster Groq for batches, plus speaker diarization—all in a polished, dark-mode RTL interface on blitz ai github.

Who should use this?

Video editors subtitling Hebrew podcasts or courses, content creators transcribing YouTube lectures, and devs building professional github profiles with open-source tools for media workflows. Ideal for anyone handling RTL languages who wants offline privacy without Descript's $24/mo lock-in.

Verdict

Grab it if Hebrew transcription is your jam—solid docs, quick start script, and MIT license make setup painless despite 47 stars and 1.0% credibility score signaling early maturity. Polish the professional github readme for your own fork; it's production-ready for solos but needs more tests for teams.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.