> ## Documentation Index
> Fetch the complete documentation index at: https://docs.videodraft.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Speech & Voice Models

> Natural text-to-speech with 60+ voices plus voice cloning

## Overview

VideoDraft provides 60+ professional voices from Google, OpenAI, and ElevenLabs, plus the ability to clone your own voice. Create natural-sounding narration in multiple languages with unlimited speech generation on paid plans.

<Info>
  Pro, Visionary, Studio, and Crew plans include unlimited speech generation — no credits consumed for voiceovers!
</Info>

## Voice Providers

### Provider Comparison

| Provider         | Voices | Cost           | Best For        |
| ---------------- | ------ | -------------- | --------------- |
| Google Wavenet   | 8+     | 1 cr/100 chars | Natural TTS     |
| Google Chirp3-HD | 4+     | 1 cr/100 chars | HD quality      |
| OpenAI TTS       | 9      | 1 cr/100 chars | Expressive AI   |
| ElevenLabs       | 42+    | 2 cr/100 chars | Premium quality |
| Voice Cloning    | Custom | Included       | Your voice      |

<Note>
  Paid plans include unlimited speech generation, so these costs only apply to Starter plan users.
</Note>

## Google Voices

### Wavenet Voices

High-quality neural network voices with natural prosody.

<Tabs>
  <Tab title="English">
    <CardGroup cols={2}>
      <Card title="Alex (Male)">
        Voice ID: en-US-Wavenet-D

        * Natural American accent
        * Professional tone
        * Clear articulation
        * Versatile delivery
      </Card>

      <Card title="Maria (Female)">
        Voice ID: en-US-Wavenet-C

        * Warm, friendly tone
        * American accent
        * Engaging delivery
        * Perfect for education
      </Card>
    </CardGroup>
  </Tab>

  <Tab title="Spanish">
    <CardGroup cols={2}>
      <Card title="Carlos (Male)">
        Voice ID: es-ES-Wavenet-B

        * Native Spanish speaker
        * Clear pronunciation
        * Professional tone
      </Card>

      <Card title="Lucia (Female)">
        Voice ID: es-ES-Wavenet-C

        * Warm Spanish voice
        * Natural intonation
        * Engaging style
      </Card>
    </CardGroup>
  </Tab>

  <Tab title="Hindi">
    <CardGroup cols={2}>
      <Card title="Ravi (Male)">
        Voice ID: hi-IN-Wavenet-B

        * Natural Hindi voice
        * Clear diction
        * Professional delivery
      </Card>

      <Card title="Sita (Female)">
        Voice ID: hi-IN-Wavenet-A

        * Warm Hindi voice
        * Natural flow
        * Expressive tone
      </Card>
    </CardGroup>
  </Tab>

  <Tab title="Telugu">
    <CardGroup cols={2}>
      <Card title="Raju (Male)">
        Voice ID: te-IN-Chirp3-HD-Achird

        * Natural Telugu voice
        * HD quality
        * Clear pronunciation
      </Card>

      <Card title="Lakshmi (Female)">
        Voice ID: te-IN-Chirp3-HD-Achernar

        * Telugu HD voice
        * Natural intonation
        * Professional quality
      </Card>
    </CardGroup>
  </Tab>
</Tabs>

### Chirp3-HD Voices

Next-generation HD quality voices with enhanced naturalness.

* Enhanced clarity and pronunciation
* Better emotional expression
* Regional accent support
* Improved prosody

## OpenAI TTS Voices

9 distinct AI voices with unique personalities:

<CardGroup cols={3}>
  <Card title="Alloy">
    Female — Balanced, clear
    Versatile voice for general use
  </Card>

  <Card title="Ash">
    Male — Mature, sophisticated
    Professional, authoritative tone
  </Card>

  <Card title="Ballad">
    Female — Smooth, melodic
    Storytelling and narration
  </Card>

  <Card title="Coral">
    Female — Vibrant, lively
    Energetic, engaging content
  </Card>

  <Card title="Echo">
    Male — Calm, thoughtful
    Meditative, educational content
  </Card>

  <Card title="Fable">
    Male — Warm, storyteller
    Narratives and podcasts
  </Card>

  <Card title="Nova">
    Female — Bright, energetic
    Marketing and promotional
  </Card>

  <Card title="Onyx">
    Male — Deep, authoritative
    Corporate and serious content
  </Card>

  <Card title="Sage">
    Female — Wise, contemplative
    Educational and thoughtful
  </Card>
</CardGroup>

<Card title="Shimmer">
  Female — Soft, gentle
  Calm, soothing content
</Card>

## ElevenLabs Voices

42+ premium voices with extensive range:

### Voice Categories

<Tabs>
  <Tab title="Professional">
    * Adam — Professional male narrator
    * Mark — Corporate presentations
    * Sandra — Business female voice
    * James — Authoritative announcer
    * Laura — Clear female presenter
  </Tab>

  <Tab title="Character">
    * Archie — Friendly character voice
    * Spuds Oxley — Distinctive personality
    * Dr. Von — Expert/scientist type
    * Northern Terry — Regional character
    * Bradford — British accent
  </Tab>

  <Tab title="International">
    * Krishna — Indian accent
    * Priyanka — Indian female
    * Viraj — South Asian male
    * Anika — German accent
    * Célian — French accent
    * Taksh — Asian voice
  </Tab>

  <Tab title="Creative">
    * Juniper — Youthful energy
    * Eve — Mysterious tone
    * Brittney — Casual American
    * Hope — Optimistic tone
    * Blondie — Bright personality
    * Arabella — Elegant voice
  </Tab>
</Tabs>

### ElevenLabs Quality

* Premium neural synthesis
* Extensive emotional range
* Character-specific voices
* Multiple accents available
* Best for professional productions

## Voice Cloning

Create custom voices from your own recordings.

### How It Works

<Steps>
  <Step title="Record Samples">
    Upload 1-5 minutes of clear speech audio
  </Step>

  <Step title="AI Learning">
    Our AI analyzes speech patterns, tone, and characteristics
  </Step>

  <Step title="Voice Created">
    Your custom voice is ready for any text
  </Step>

  <Step title="Use Anywhere">
    Generate unlimited speech with your cloned voice
  </Step>
</Steps>

### Recording Requirements

For best results:

* Quality: Clear audio, minimal background noise
* Length: 1-5 minutes of varied speech
* Content: Natural conversation, varied sentences
* Format: MP3, WAV, or M4A

### Voice Cloning Tips

<AccordionGroup>
  <Accordion title="Recording Best Practices">
    * Use a quality microphone
    * Record in a quiet environment
    * Speak naturally, not too fast
    * Include varied intonation
    * Avoid heavy processing/effects
  </Accordion>

  <Accordion title="Improving Clone Quality">
    * Provide more sample audio
    * Include different emotional tones
    * Ensure consistent audio quality
    * Re-record if results are poor
  </Accordion>
</AccordionGroup>

### Plan Availability

| Feature          | Starter | Pro | Visionary | Studio    | Crew      |
| ---------------- | ------- | --- | --------- | --------- | --------- |
| Voice Cloning    | —       | Yes | Yes       | Yes       | Yes       |
| Clone Storage    | —       | 3   | 10        | Unlimited | Unlimited |
| Unlimited Speech | —       | Yes | Yes       | Yes       | Yes       |

## Voice Selection Guide

### By Content Type

<CardGroup cols={2}>
  <Card title="Corporate/Professional">
    Recommended:

    * Alex (Google) — Authoritative male
    * Onyx (OpenAI) — Deep, professional
    * Adam (ElevenLabs) — Premium corporate
    * Mark (ElevenLabs) — Clear presenter
  </Card>

  <Card title="Educational">
    Recommended:

    * Maria (Google) — Friendly, clear
    * Sage (OpenAI) — Thoughtful, wise
    * Echo (OpenAI) — Calm, measured
    * Nova (OpenAI) — Engaging, energetic
  </Card>

  <Card title="Marketing">
    Recommended:

    * Nova (OpenAI) — Bright, energetic
    * Coral (OpenAI) — Vibrant, lively
    * Alloy (OpenAI) — Versatile
    * ElevenLabs characters — Unique personality
  </Card>

  <Card title="Storytelling">
    Recommended:

    * Fable (OpenAI) — Natural storyteller
    * Ballad (OpenAI) — Melodic, flowing
    * ElevenLabs characters — Distinctive voices
    * Voice clones — Personal touch
  </Card>
</CardGroup>

### By Language

| Language     | Google        | OpenAI       | ElevenLabs  |
| ------------ | ------------- | ------------ | ----------- |
| English (US) | Alex, Maria   | All 9 voices | 30+ options |
| English (UK) | Available     | Some         | 10+ options |
| Spanish      | Carlos, Lucia | Limited      | Available   |
| Hindi        | Ravi, Sita    | —            | Limited     |
| Telugu       | Raju, Lakshmi | —            | —           |
| French       | Available     | —            | Célian +    |
| German       | Available     | —            | Anika +     |

## Script Writing for TTS

### Best Practices

<Tabs>
  <Tab title="Clarity">
    Write for the ear:

    * Use simple sentences
    * Avoid complex punctuation
    * Break up long thoughts
    * Use natural pauses

    Example:

    ```
    Instead of: "The product—which launched in 2023—has received numerous accolades."

    Write: "The product launched in 2023. It has received numerous accolades."
    ```
  </Tab>

  <Tab title="Pacing">
    Control speech rhythm:

    * Period = full stop (0.5-1s pause)
    * Comma = brief pause (0.2s)
    * Ellipsis... = longer pause
    * New paragraph = topic change
  </Tab>

  <Tab title="Emphasis">
    Guide inflection:

    * CAPS for strong emphasis
    * Questions for engagement
    * Exclamations for energy
    * Vary sentence length
  </Tab>
</Tabs>

### Pronunciation Tips

Numbers and Dates:

* "2024" → "twenty twenty-four"
* "\$99" → "ninety-nine dollars"
* "3/4" → "three quarters"

Abbreviations:

* "Dr." → "Doctor"
* "vs." → "versus"
* Spell out acronyms: "NASA (nah-sah)"

## Integration

### With Avatar Video

Speech synthesis powers Avatar Video:

1. Write script
2. Select voice
3. AI generates speech
4. Lip sync applied to avatar
5. Complete talking head video

### With Storyboard

Add narration to any scene:

1. Write scene narration
2. Select voice per scene
3. Generate speech
4. Audio syncs to visuals

### With AI Film Crew

The Screenwriter agent helps write TTS-optimized scripts:

* Natural dialogue
* Appropriate pacing
* Emotional delivery cues
* Clear pronunciation notes

## Next Steps

<CardGroup cols={3}>
  <Card title="Voice Library" icon="microphone" href="https://app.videodraft.ai/voice-library">
    Manage your voices
  </Card>

  <Card title="Avatar Video" icon="user-circle" href="/features/avatar-video">
    Create talking avatars
  </Card>

  <Card title="How Credits Work" icon="coins" href="/how-credits-work">
    Understand pricing
  </Card>
</CardGroup>

<Tip>
  Pro tip: Test 2-3 voices with a short script before committing to one for your entire project. Different voices suit different content styles!
</Tip>