Overview
ElevenLabs audio tags are control commands wrapped in square brackets[tag] that direct the AI voice’s emotional delivery, tone, and performance. These tags are placed inline with your text and are not spoken aloud—instead, they modify how the surrounding text is delivered.
How to Use Audio Tags
- Place tags anywhere in your text:
"Hello there [whispers] can you hear me?" - Combine multiple tags:
"[nervously][whispers] I think someone's coming" - Match tags to your voice’s character for best results
Available Audio Tags
Emotional Expression
| Tag | Description |
|---|---|
[happy] | Cheerful, upbeat delivery |
[sad] | Melancholy, mournful tone |
[excited] | Energetic, enthusiastic delivery |
[angry] | Irritated, upset tone |
[nervous] | Anxious, worried delivery |
[calm] | Peaceful, relaxed tone |
[frustrated] | Annoyed, exasperated delivery |
[curious] | Inquisitive, questioning tone |
[tired] | Weary, exhausted delivery |
[mischievously] | Playful, scheming tone |
Vocal Delivery Styles
| Tag | Description |
|---|---|
[whispers] | Quiet, intimate delivery |
[shouts], [shouting] | Loud, emphatic delivery |
[sarcastic], [sarcastically] | Ironic, mocking tone |
[dramatic] | Theatrical, intense delivery |
[matter-of-fact] | Straightforward, neutral tone |
[cheerfully] | Bright, pleasant delivery |
[flatly] | Monotone, emotionless delivery |
[deadpan] | Dry, expressionless tone |
[playfully] | Fun, lighthearted delivery |
[whiny] | Complaining, petulant tone |
Laughter & Amusement
| Tag | Description |
|---|---|
[laughs] | Natural laughter |
[laughs harder] | More intense laughter |
[starts laughing] | Beginning to laugh |
[giggles] | Light, playful laughter |
[light chuckle] | Subtle amusement |
[wheezing] | Breathless, intense laughter |
[snorts] | Snorting laugh |
Physical Reactions & Sounds
| Tag | Description |
|---|---|
[sighs] | Exhale of resignation or relief |
[exhales] | Breathing out |
[gasps], [gasp] | Sharp intake of breath (surprise/shock) |
[gulps] | Nervous swallowing |
[swallows] | Swallowing sound |
[clears throat] | Attention-getting throat clearing |
[crying] | Tearful, sobbing delivery |
Cognitive States
| Tag | Description |
|---|---|
[pauses] | Brief hesitation |
[hesitates] | Uncertain, wavering delivery |
[stammers] | Stuttering, nervous speech |
[resigned tone] | Accepting, giving up |
[regretful] | Sorry, remorseful tone |
[hesitant] | Unsure, cautious delivery |
Character Voices & Accents
| Tag | Description |
|---|---|
[French accent] | French-accented delivery |
[Australian accent] | Australian-accented delivery |
[strong X accent] | Replace X with desired accent |
[pirate voice] | Pirate character voice |
[fantasy narrator] | Fantasy storytelling voice |
[sci-fi AI voice] | Futuristic AI character |
[classic film noir] | 1940s detective style |
Sound Effects
| Tag | Description |
|---|---|
[applause] | Clapping sounds |
[clapping] | Hand clapping |
[gunshot] | Gunshot sound effect |
[explosion] | Explosion sound effect |
Special & Experimental
| Tag | Description |
|---|---|
[sings] | Singing delivery |
[woo] | Exclamation of excitement |
[dismissive] | Brushing off, uninterested |
[cheeky] | Impudent, bold tone |
Best Practices
Tag Placement
- Use tags at natural speech breaks
- Place emotional tags before the text they should affect
- Sound effect tags can be placed mid-sentence for realistic timing
Voice Compatibility
- Test tags with your chosen voice—effectiveness varies by voice character
- Serious voices work better with professional tags
- Playful voices respond well to emotional and humorous tags
Combining Tags
- Stack multiple tags for complex emotions:
[tired][sighs] - Transition between emotions:
[excited] Great news! [pause] [serious] But we need to talk.
Experimental Tags
- Many tags beyond this list may work—experiment with descriptive emotional states
- Test thoroughly before production use as experimental tags may be inconsistent