Discussions

When my LiveAvatar starts speaking, I want it to read, verbatim, and introduction paragraph. When I prompt it to, it responds, "I cannot say that" or somesuch. How do I prompt the avatar to speak my text? I see "opening_text" in API docs, but the API docs don't really explain any of the parameters, so I don't know what that is.

Hi everyone,

My avatar doesnt respond back to me, I start listening it listens it shows caption like transcription but I get 0 response.

I read through your docs and am somewhat confused. I am building an assistant for our college, and they want a virtual assistant as well. Right now, we have our own LLM but need to integrate it with your avatar. FULL mode is not an option since we have our own LLM. I was wondering if when using LITE mode, there is any way to get the voice for the avatar from you. In the docs, it says we have to generate it? Also, it says the API only accepts audio files, is there any way we can just pass the responses from our LLM, which has TTS from the user side already implemented into your model?

When trying to integrate with a voice agent, I'm encountering a problem. Providers, such as Vapi or other companies, continuously stream the audio output. However, in the liveavatar SDK, I only have the repeatAudio function. Unfortunately, providing this function with small chunks of the streamed audio isn't enough. So, what can I do to solve this problem? Trying to understand how the FULL mode works, I suspect you're also performing a streaming operation. Because no matter how long the generated text is, the avatar still starts speaking after a short time. How can I solve this problem? How did you solve it?

I have created a custom live avatar. I want to cancel the subscription for a period of time and maybe rejoin when I need it again. If I cancel the subscription, would my created custom live avatar be deleted forever or I can use it once I subscribe again?

I’m currently using heygen/liveavatar-web-sdk version ^0.0.10 with Next.js.

{
"code": 4002,
"message": "Invalid token: Not enough segments"
}

Hey guys,

Problem:
Custom LiveAvatar mode supports ElevenLabs voice integration but lacks real-time interruption handling, unlike full mode.