BeltoVox AI Transcription

BeltoVox User Guide

Version 1.0

Documentation

🎙️ BeltoVox User Guide & Site Map

1. Getting Started (Setup Guide)

Use this section for your "Installation" or "Setup" page.

Setting up your AI Engines

BeltoVox connects directly to the world’s most powerful AI models. To begin, you need to provide your own API keys.

Groq API (Recommended)

The fastest transcription engine available.

  • Visit console.groq.com.
  • Create a free API Key.
  • Paste it into the BeltoVox Dashboard.

OpenAI API

The industry standard for accuracy.

  • Visit platform.openai.com.
  • Add a small credit balance (e.g., $5).
  • Create a secret key and add it to BeltoVox.

2. The Main Dashboard

This section explains the main settings window where all configuration happens.

Tab 1: API Settings

Purpose: Manages your connection to the AI "brains."

Controls:

  • Switch between Groq and OpenAI.
  • Securely store your API keys (encrypted locally).
Best Practice: Keep both keys saved so you can switch if one provider is down.

Tab 2: Triggers & Output

Purpose: Customizes how you start dictating and where the text goes.

Controls:

  • Keyboard Hotkey: Set a global shortcut (e.g., Ctrl+Space) to record anywhere.
  • Mouse Triggers: Assign recording to your Mouse Side Buttons or Middle Click.
  • Delivery Mode: Choose "Direct Insert" to have text typed at your cursor, or "History Only" to save it for later.

Tab 3: AI & Language

Purpose: Fine-tunes the "intelligence" of your dictation.

Controls:

  • Smart AI Modes: Choose between Raw (Literal), Refiner (fixes grammar), Professional (formal), or Friendly (warm).
  • Translation: Toggle "Translate to English" to speak in another language and get English text back.
  • Language Selection: Manually set your language or use Auto-Detect.

Tab 4: Audio

Purpose: Manages your hardware and feedback.

Controls:

  • Device Selector: Choose which microphone to use.
  • Quality Settings: Adjust sample rates for better accuracy.
  • Sound Cues: Enable or disable the "Beeps" that signal recording start/stop.

Tab 5: History

Purpose: Your personal archive of everything you've said.

Controls:

  • Scroll through past transcriptions.
  • One-click "Copy All" or "Clear History."
  • Right-click specific entries to copy just that segment.

Tab 6: Usage & Stats

Purpose: Transparent tracking of your productivity and costs.

Controls:

  • Monitor total recording time.
  • Track API requests.
  • View an Estimated Cost tracker to see exactly what you are spending on AI.

3. Quick Control Panel

Use this for a "Minimalist Interface" or "Workflow" section on your site.

What it is: A compact, always-on-top window accessed by left-clicking the tray icon.

Functions:

  • Instant Controls: Big, clickable Record, Stop, and Pause buttons.
  • 👁️ Preview Window: See your text as it arrives without opening the full dashboard.
  • 🌐 Quick Lang Switch: Change your input language on the fly.
  • 🔄 View Switcher: Move between "Preview" (current) and "History" (recent) views.

4. Visual Status Overlay

This is one of your "Premium Features" to highlight.

The Floating Indicator: A semi-transparent status circle that stays on top of your work.

Gray: Idle and ready
Red: Recording (Live Mic)
Blue: Processing (AI Thinking)
Yellow: Paused

5. System Tray & Hotkeys

This section explains the "invisible" power of the app.

Background Operation

BeltoVox lives in your tray (near the clock). It stays out of your way until you need it.

The Global Hotkey

By default, Ctrl + Space triggers the mic. No need to click any buttons—just press, talk, and release.

Fast Tray Menu: Right-click the tray icon for instant access to Language settings and App exit.

6. Security & Encryption

Critical for building trust on your website.

🔒

Local Encryption

Your API keys are encrypted with Fernet symmetric encryption using a machine-specific key. No one else can read them. They never leave your device.

🛡️

Privacy First

We don't store your audio. Files are processed in memory or as temporary files and deleted the millisecond the transcription is done.