NanoMaker Docs

Text to Speech

Transform text into natural speech using 26 professional ElevenLabs AI voices across 5 categories. Perfect for voiceovers, narration, and podcasts. 10 credits per 1K chars.

Overview

Transform written text into natural-sounding speech using ElevenLabs' professional AI voice technology. Perfect for voiceovers, narration, and audio content.

Text to Speech

Cost: 10 credits per 1,000 characters
Voices: 26 professional options in 5 categories
Time: ~30 seconds
Format: High-quality MP3

Natural, human-like speech powered by ElevenLabs Turbo 2.5.

How to Use

Select Text to Speech

Navigate to the AI Generator and select the Text to Speech tab.

Enter Your Text

Type or paste the text you want to convert to speech (max 5,000 characters).

Choose a Voice

Click the voice selector and preview different voices by clicking the speaker icon next to each name. Voices are organized into 5 categories for easy selection.

Generate

Click "Generate" - cost is calculated automatically based on character count.

Available Voices

We offer 26 professional voices organized into 5 categories:

Standard Voices (10)

Professional, versatile voices suitable for most content types.

Audiobook (1)

Specialized voice for long-form narration and storytelling.

Radio Hosts (4)

Dynamic, engaging voices perfect for broadcasting and entertainment.

TikTok & Social Media (6)

Trendy, relatable voices optimized for social media content.

Video Games (5)

Character voices perfect for gaming content and entertainment.

Preview Before Generating

Click the speaker icon next to each voice name in the selector to hear a sample before generating!

Pricing

Text to Speech uses character-based pricing:

CharactersCredits
1-1,00010
1,001-2,00020
2,001-3,00030
3,001-4,00040
4,001-5,00050

Formula: Credits = Math.ceil(characters / 1000) × 10

Limits:

  • Minimum: 1 character (10 credits)
  • Maximum: 5,000 characters per generation

The credit cost updates in real-time as you type your text.

Writing for Speech

Text written for reading doesn't always work well for speech. Follow these tips:

Best Practices

Write as You Speak

❌ Written style:

The aforementioned solution provides optimal results.

✅ Spoken style:

This solution works really well.

Use Contractions

  • "don't" instead of "do not"
  • "we're" instead of "we are"
  • "it's" instead of "it is"

Keep Sentences Short

  • Break long sentences into shorter ones
  • Each sentence should have one main idea
  • Use simple, clear language

Punctuation Controls Pacing

Periods (.): Natural pause

Welcome to our podcast. Today we're discussing AI technology.

Commas (,): Brief pause

First, let's review the basics. Then, we'll dive deeper.

Question Marks (?): Rising intonation

What do you think? Isn't this amazing?

Exclamation Marks (!): Emphasis and energy

That's incredible! Let's celebrate!

Em Dashes (—): Longer, dramatic pause

And then—after years of work—we finally succeeded.

Clean Text Works Best

Remove:

  • Markdown syntax (**bold**, *italic*)
  • HTML tags (<strong>, <em>)
  • Special characters
  • URLs (spell them out or describe them)
  • Code snippets

Keep:

  • Natural punctuation
  • Numbers (spelled out for clarity)
  • Abbreviations (if commonly spoken)

❌ Avoid:

Check out [our website](https://example.com) for **more info**!

✅ Use:

Check out our website for more information!

Numbers & Abbreviations

Numbers:

  • Write "five hundred" instead of "500" for clearer pronunciation
  • Use digits only if that's how you'd say it: "2024", "iPhone 15"

Abbreviations:

  • "AI" → pronounced as letters
  • "NASA" → pronounced as word
  • "Dr." → "Doctor"

Use Cases

Quality Tips

Preview Voices

Always preview voices before generating. Each voice has a unique character that works better for different content types.

Optimize Text

  • Break very long texts into smaller chunks
  • Remove special formatting
  • Use natural punctuation for pacing
  • Write how you would actually speak

Character Limit

Maximum 5,000 characters per generation. For longer content, split into multiple generations and edit together.

Common Issues

Next Steps