Skip to content

AI-powered desktop app for large-scale transcription of audio, images, and PDFs into high-quality text, with built-in quality checks and remediation.

License

Notifications You must be signed in to change notification settings

Minitex/TranscribeAIUI

Repository files navigation

TranscribeAI Logo

TranscribeAI turns audio files and image scans into text quickly and accurately.

Built for a simple setup and high-volume transcription workflows.

Get latest TranscribeAI version Total TranscribeAI downloads across GitHub release assets


Stay Tuned

⭐️ Star our repository to stay up-to-date with exciting new features and improvements! Get instant notifications for new releases!

How to star this repository

What You Can Do

  • Transcribe audio files to text.
  • Transcribe images and PDFs to text.
  • Run a Quality Check to flag possible issues.
  • Use Remediate for safe cleanups (intro/outro chatter, markdown artifacts, encoded entities).
  • Get clear warnings for subtitle timestamp issues so you can review those sections manually.
  • Track everything in real-time with progress and logs.

Before You Start

You need at least one API key:

Gemini vs Mistral (Quick Guide)

  • Mistral: Very fast, especially when processing large folders in batches. Audio results are also very good.
  • Gemini: Slightly better image accuracy in many cases, so it is a strong OCR choice when precision matters most.
  • Many users add both keys: use Mistral when speed matters and Gemini when image accuracy matters most.

More AI model support is coming soon.

Install

  1. Open the latest release.
  2. Download your installer:
    • macOS: .dmg
    • Windows: .exe
    • Linux: .AppImage or .tar.gz
  3. Run the installer.

If your system shows a first-run security warning:

  • macOS: open Privacy/Security and choose Open Anyway
  • Windows: choose More info then Run anyway

First-Time Setup

  1. Open TranscribeAI.
  2. Click the gear icon (top-right) to open Settings.
  3. Paste your API key(s).
  4. Click Save.

Basic Use

  1. Choose mode: Audio or Image.
  2. Select input file/folder.
  3. Select output folder.
  4. Click Transcribe.
  5. Watch progress in the logs.

Quality Check

  • Click Check Quality to score your transcript/subtitle files.
  • Lower scores mean the app found more potential issues.
  • Hover a score to see the penalty breakdown (why it scored lower).
  • Red warning dots mean “review needed.”

For subtitle files (.srt):

  • The app flags timestamp format/order/range issues.
  • It shows cue numbers so you can quickly find where to listen and fix.
  • Timestamp fixes are manual review (not auto-adjusted).

Updating

  • Use the latest release.
  • Install over your current version.
  • Your existing settings and data remain.

License

This project uses the MIT License. See LICENSE.

Screenshot

TranscribeAI Main Interface

About

AI-powered desktop app for large-scale transcription of audio, images, and PDFs into high-quality text, with built-in quality checks and remediation.

Resources

License

Stars

Watchers

Forks

Contributors 2

  •  
  •  

Languages