Pronunciation Guides

Pronunciation Guides help you control how the AI narrator pronounces specific words, brand names, technical terms, and acronyms in your videos. This ensures consistent, accurate narration across all your projects.

What are Pronunciation Guides?

Pronunciation Guides are reusable collections of pronunciation rules that you can apply to your projects. Each guide contains rules that map words or phrases to their phonetic pronunciation.

For example:

  • SQL → sequel - Pronounce "SQL" as "sequel" instead of "S-Q-L"
  • API → A-P-I - Spell out each letter
  • AiToVid → AI to vid - Break down your brand name

Creating a Pronunciation Guide

  1. Navigate to Pronunciation Guides from the main menu
  2. Click "New Guide"
  3. Fill in the guide details:
    • Name (required) - A descriptive name for your guide (e.g., "Tech Terms" or "Company Branding")
    • Description (optional) - Additional context about this guide
    • Content (optional) - Legacy text-based pronunciation rules
  4. Click "Create Guide"

Managing Pronunciation Rules

Once you've created a guide, click "Manage Rules" to add, edit, or organize pronunciation rules.

Adding Individual Rules

  1. Click "Add Rule"
  2. Fill in the rule details:
    • Input - The word or phrase as it appears in your script (e.g., "SQL")
    • Output - How it should be pronounced (e.g., "sequel")
    • Notes (optional) - Additional context or reminders about this rule
  3. Click "Create Rule"

Bulk Adding Rules

For faster setup, you can add multiple rules at once:

  1. Click "Bulk Add"
  2. Enter rules in the format: input → output
    • Each rule should be on a new line
    • You can use →, ->, or => as separators
    • Example: 
      SQL → sequel
API → A-P-I
AiToVid → AI to vid
  3. Click "Create Rules"

Rule Format

Rules use simple text replacement:

  • Input: The text to match (case-insensitive)
  • Output: The replacement text for pronunciation
  • Notes: Optional documentation for your team

The system uses global, case-insensitive matching, so "SQL", "sql", and "Sql" will all be replaced.

Rule Priority and Order

Rules are applied in priority order:

  • Lower priority numbers are applied first
  • You can reorder rules by dragging them or using priority controls
  • Rules are applied sequentially, so earlier rules can affect later ones

Testing Your Pronunciation Guide

Before using your guide in a project, you can test how it will sound:

  1. In the guide detail page, find the "Preview & Test" section
  2. Enter sample text that includes your pronunciation terms
  3. Click "Preview" to see how the text will be transformed
  4. Click "Test Audio" to hear how it will sound with AI narration
  5. Adjust your rules based on the results

Using Pronunciation Guides in Projects

When Creating a New Project

  1. In the "New Project" modal, find the "Pronunciation Guide" section
  2. Choose one of three options:
    • None - No pronunciation guide
    • Use existing - Select from your saved guides
    • Custom - Enter one-time pronunciation rules for this project only

Updating an Existing Project

  1. Open your project
  2. Click "Edit Project" in the action bar
  3. Change the pronunciation guide settings
  4. Regenerate scripts or audio to apply the changes

Guide Status

Guides can have two status states:

  • Active - Available for use in projects
  • Archived - Hidden from project selection but preserved for projects already using them

To archive a guide, click the "Archive" button on the guide card. You can reactivate it anytime.

Duplicating Guides

To create variations of existing guides:

  1. Find the guide you want to copy
  2. Click "Duplicate"
  3. The new guide will be named "[Original Name] (Copy)"
  4. Edit the duplicate to make your changes

Deleting Guides

You can delete guides that are no longer needed:

  • Guides in use by projects cannot be deleted
  • The guide card shows usage count (e.g., "5 uses")
  • Archive unused guides instead of deleting them for better record-keeping

Best Practices

Start Simple

  • Begin with your most commonly mispronounced terms
  • Test each rule before adding complex ones
  • Build your guide incrementally

Use Descriptive Names

  • Name guides by category (e.g., "Medical Terms", "Product Names")
  • Add descriptions to help team members understand when to use each guide

Test Before Production

  • Always test pronunciation rules with the audio preview
  • Check how rules interact with different contexts
  • Verify that rules don't affect unintended words

Organize by Project Type

  • Create separate guides for different industries or clients
  • Reuse guides across similar projects for consistency
  • Share guides with team members through project collaboration

Phonetic Spelling Tips

  • Use hyphens to spell out acronyms: API → A-P-I
  • Use phonetic spelling for complex words: acetaminophen → a-seat-a-min-o-fen
  • Test with the audio preview to fine-tune pronunciation

Exporting Guides

You can export pronunciation guides for:

  • Backup purposes
  • Sharing with team members
  • Migration to other systems

The export feature is available from the guide detail page.

How Pronunciation Rules Are Applied

When generating audio for your project:

  1. The system retrieves your selected pronunciation guide
  2. All enabled rules are sorted by priority
  3. Each rule is applied to the script text in sequence
  4. The transformed text is sent to the text-to-speech engine
  5. The AI narrator uses the modified text for pronunciation

This happens automatically whenever you generate or regenerate audio for slides.