Discussions
Any timeline for the following features?
- Arabic support
- Multilingual avatars (automatically switch language based on user's language)
- Custom avatar using a photo or AI-generated video.
Text payload/conversation integration with existing LLM
We have an existing LLM that we want to use with a live avatar. So we want a custom avatar, a user will ask our interface a question and then the response from our LLM will be sent to the Live Avatar as a text payload which will then speak it. Is this possible with the custom Avatar?
Avatar Personalizado
Existe la posibilidad de personalizar el avatar o solo se puede elegir de la lista que esta disponible actualmente
Creating avatars for others
Heygen supports multiple persons to create avatars in a single account (Team Plan). If I have I understood correctly: Heygen and LiveAvatar are different entities (platforms do not discuss, cannot import/export avatars, and so on).
Getting final video
Given a live conversation with the avatar, is it possible to download a video of the entire conversation?
Botpress Integration
We currently use Botpress for our chatbot and have built a good knowledge base and foundation. How can we integrate out Botpress chatbot (as the brains) and LiveAvatar?
Interruption Handling issue in Avatar
Hi, in the full mode and the demos available on website, the avatar should handle interruptions well and stop speaking midway a conversation if the user starts asking some question.
Background replacement
Hello, I have a use-case where I want to use one of the stock avatars, but replace the background - is this possible?
Custom Callback Events — Is This Supported?
We’d like to know if LiveAvatar supports triggering a custom callback or webhook event (from within a session) that our system can listen to (for example when a user reaches a certain point in the conversation). The aim is that based on the conversation context we could redirect the user to a specific part of our website.
How to fix "400006 - Session is not in correct state"?
We use CUSTOM mode, and when we start sending audio through the websocket (ws_url) we get frequent: