Skip to main content
expressiveness_boost sets how much emphasis, pitch movement, and rhythm a voice puts into the words. It is a number from 0.0 to 1.0 in the body of POST /v3/models/audio/tts and /tts/stream, set per request. It works with public voices and instant voices.

Listen

The same script with expressiveness_boost set to 0.0 and 1.0. Listen for emphasis, pitch movement, rhythm, and pauses.
I’ve been waiting to show you this. You know that moment when an idea finally starts to feel real? The first version is never perfect. You try again, change a few things, and suddenly it clicks. That’s the part I love. Taking something you imagined and turning it into something you can share. I can’t wait to see what you create with it.

0.0

1.0

Values

Any number in between also works.

Choose a boost level

Use higher values for more emphasis and variation in delivery, or lower values for a more restrained reading. To compare values, keep the voice, model, text, punctuation, and language the same and change only the boost. For the closest match to an instant clone’s original speaker, create the clone with similarity: "max" and send expressiveness_boost: 0.0. For HeyGen videos, our preferred combination is similarity: "max" with expressiveness_boost: 0.5.