Skip to main content
Voice settings control how your agent sounds to callers. Voice settings live on the Model tab of the agent editor, in the Agent Settings panel on the right.

​
Voice

Wave Runner offers a library of voices across multiple languages and accents. Click the play button next to the voice dropdown to preview any voice before selecting it.

​
Language

Set the Language dropdown before choosing a voice. The language setting also controls the speech-to-text transcription model. Available languages: English, Spanish, Portuguese, French, German, Dutch, Italian, Hindi, Chinese, English & Spanish.
If your callers speak both English and Spanish, select English & Spanish to handle code-switching.

​
Speaking speed

The Speed slider runs from 0.5× to 2.0×. Default is 1.0×. Slower speeds work well for complex instructions, legal disclosures, or callers who are non-native speakers. Faster speeds suit high-volume outbound calls with short, conversational exchanges.
Small adjustments of ±0.2 make a noticeable difference. Stay close to 1.0× for most use cases.

​
Response style

Response style controls how consistently the agent phrases its responses from call to call.
Focused is the right choice for most agents. It produces the most consistent, testable behavior.

​
Background sound

Background sound lives in the Advanced tab under Call behavior. It plays ambient audio during calls to reduce silence between responses.

​
Next steps

Create an agent

Full walkthrough of every setting in the agent editor.