We've shipped a round of updates focused on giving you more control over your AI phone agents and making the setup process smoother.
Onboarding Improvements
We've streamlined the onboarding flow to get you from sign-up to a live AI phone agent faster. The process is now simpler, with clearer steps and fewer friction points along the way.
Voice Agent Editor
The voice agent editor got a significant upgrade. You can now select from multiple voices when building your agent, so you can pick the one that best fits your brand and your callers' expectations.
Advanced Settings
We added a new Advanced Settings section to the voice agent editor, giving you fine-grained control over how your agent behaves:
- LLM Model Selection - Choose the language model powering your agent. We recommend sticking with the default (GPT-4.1), which works great for most use cases. Most people won't need to change this.
- Language Selection - Set the language for your calls. We still recommend writing your prompts in English for the best results, but you now have the flexibility to configure the call language.
- Background Sounds - Add ambient audio to make calls feel more natural. Options include: Coffee Shop, Convention Hall, Summer Outdoor, Mountain Outdoor, Static Noise, and Call Center.
- Response Eagerness - Control how quickly the agent starts responding after the caller finishes speaking.
- Interruption Sensitivity - Adjust how sensitive the agent is to being interrupted mid-sentence by the caller.
What's Next
We're continuing to improve the voice agent editor to make building and customizing your AI agents easier than ever. We're also working on adding more templates to our template library, so you can get started even faster. Stay tuned.