Expression Tags

Add emotional depth to your audio with expression tags.

Available Expressions

Usage Examples

Self-Contained Expressions

These tags produce a sound effect at that position:

import requests

# Laughter
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "That's hilarious! <laugh> I can't believe it!",
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-prime"
    }
)

# Chuckle
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "Well, that's funny <chuckle> but let's continue.",
        "voice": "Emma",
        "language": "en-US",
        "model": "aura-prime"
    }
)

# Sigh
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "<sigh> It's been a really long day.",
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-prime"
    }
)

# Gasp
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "<gasp> I didn't expect that at all!",
        "voice": "Emma",
        "language": "en-US",
        "model": "aura-prime"
    }
)

Wrapping Expressions

These tags modify how the enclosed text is spoken:

# Whisper
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "I have a secret. <whisper>Don't tell anyone.</whisper>",
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-prime"
    }
)

# Emphasis
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": "This is <emphasis>very important</emphasis> information.",
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-prime"
    }
)

Combining Expressions

# Multiple expressions in one text
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": """
            <sigh> I'm so tired today.
            But wait <gasp> what's that?
            <whisper>I think someone is coming.</whisper>
            <laugh> Just kidding! It was nothing.
        """,
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-max"
    }
)

Audiobook Example

Creating expressive narration for storytelling:

story_text = """
The detective looked at the evidence. <sigh> "It doesn't make sense," he muttered.

Suddenly, the door burst open! <gasp>

"I know who did it!" shouted Maria, breathless from running.

The detective <chuckle> leaned back in his chair. 
<whisper>This was going to be interesting.</whisper>
"""

response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": story_text,
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-max"  # Best quality for audiobooks
    }
)

Best Practices

• Use sparingly - Too many expressions can sound unnatural
• Match the context - Use expressions that fit the emotional tone
• Test with your content - Preview how expressions sound with your specific text
• Use Aura Max - For best expression quality, use aura-max model

---

Break Tag — Timed Silence

Insert precise pauses anywhere in your audio using the <break time="..."/> tag. Control pacing, add dramatic effect, or cleanly separate sections.

Syntax

Basic Usage

import requests

# 1-second pause between sentences
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": 'Welcome to YourVoic. <break time="1s"/> Let\'s get started.',
        "voice": "Peter",
        "language": "en-US",
        "model": "aura-prime"
    }
)

# 500ms pause using milliseconds
response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": 'Step one: open the app. <break time="500ms"/> Step two: sign in.',
        "voice": "Emma",
        "language": "en-US",
        "model": "aura-prime"
    }
)

Multiple Breaks

response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": """
            Welcome to the evening news.
            <break time="1s"/>
            Our top story: a major breakthrough in renewable energy.
            <break time="500ms"/>
            Scientists say this could change everything.
            <break time="2s"/>
            More details after the break.
        """,
        "voice": "Alan",
        "language": "en-US",
        "model": "aura-max"
    }
)

Combining with Expression Tags

response = requests.post(
    "https://yourvoic.com/api/v1/tts/generate",
    headers={"X-API-Key": "your_api_key"},
    json={
        "text": '<sigh> It\'s been a long day. <break time="1s"/> But we made it! <laugh>',
        "voice": "Emma",
        "language": "en-US",
        "model": "aura-prime"
    }
)

How It Works

• Text is split at each <break> tag into segments
• All text segments are synthesized in parallel for fast response
• Precise silence of the exact requested duration is inserted between each segment
• All segments are stitched into a single continuous audio file in the correct order
• If any segment fails after retries, the full request fails — you will never receive incomplete audio