​
General

​
What is the difference between an Avatar and an Agent?

An Avatar is a virtual replica of a person that is used as a base for an Agent. An Agent is an Avatar with an assigned job description and knowledge base that can carry out conversations and interactions in the digital world.

​
Billing

​
How is API usage calculated and billed?

Beyond Presence uses a Credits system, where 1 Credit is equivalent to 1 minute of active conversation. You will not be charged for the time an agent is silent or listens to the user. Further, end-to-end agents will time out after 5 minutes of inactivity and the meeting will end.

​
What pricing plans are available?

Beyond Presence offers flexible pricing plans: Free, Personal, Business and Enterprise. Visit our Pricing Page for more information.

​
How can I change my pricing plan?

In the Beyond Presence Dashboard, click on Settings > Billing > Update Subscription and select the pricing plan that fits your needs.

​
How do I monitor usage?

You can view your current usage in the Beyond Presence Dashboard under Settings > Billing.

​
Can I start for free?

Yes, you can create and test an Agent for one minute for free with our Free plan.

​
Customization

​
How do I handle multi-lingual conversations?

Beyond Presence avatars and agents natively support 35 of the most popular languages. See our Supported Languages for more information.

​
How can I create my own personal avatar?

See our Custom Avatars Guide for more information.

​
How do I integrate my own LLM?

If you have a custom LLM, we recommend using our Audio-to-Video API. Custom LLM support for the end-to-end API is also on our Roadmap and will be supported soon.

​
How do I integrate my own TTS?

If you have need to have full control over your TTS, or if you have already built an audio agent pipeline, we recommend using our Audio-to-Video API. Custom TTS support for the end-to-end APIs is also on our Roadmap and will be supported soon.

​
Optimization

​
What causes latency and how can I optimize it?

Latency can come from several sources:
  • Transcription / STT processing
  • LLM processing time
  • TTS generation
  • Avatar video creation
  • Network conditions
To optimize latency:
  • Optimize your infrastructure deployment to reduce physical distance between services
  • Prompt your LLM to respond in short and concise answers
  • Consider using our end-to-end API instead of a custom LLM/TTS for lowest latency