Discussions
Language switcher option in SDK integration NextJS
I’d like to integrate an interactive avatar into my Next.js website and allow users to select their preferred language before starting. What’s the best approach to implement this? Do you have any code samples? Also, which speech-to-text (STT) engine would you recommend to support Hindi and other Indian languages?
Using HeyGen api - Video is stuck at first frame
I've had an app hosted on Railway for two months without any problems: Public Avatar and voice.
Task Type of Chat Stoppped Working
In the streaming avatar API, using the task type of 'Chat'. has stopped working. The task type of 'repeat' still works. Please let me know if you or any other users are seeing the same issue as me.
Can I specify photo when I use "Generate photo avatar photos" API?
Here https://docs.heygen.com/reference/generate-photo-avatar-photos
you tell that I can "effortlessly create personalized AI avatars from photos".
Any way to create video avatars via API?
The problem is that I need voice ID for https://docs.heygen.com/reference/create-avatar-iv-video
and it's not possible to create a voice ID via API.
I have a lot of voices on ElevenLabs but their automatic integration via API is also not possible, it can be done only manually.
GPT 5 mini for streaming avatars?
I want to switch to GPT 5 mini for many reasons, cost, speed, more context, etc.
Brand glossary
You urgently need to make the Brand Glossary available for API use... it's clearly a tool meant to be used in mass production, but there's no way to automate it with an API... only by using the website manually.
Newbie Question
Hey there i need help creating video avatars using the api. Here's what we want to do.
Compartilhar conversas
Tenho um restaurante estou implantando um garçom inteligente mas preciso compartilhar as conversas em texto em tempo real pra cozinha
Problem with voice grounded replies
Hello, I have modified the InteractiveAvatarNextJSDemo code so that it uses data from my Azure container and vector search. This is working great for text chatting with the avatar. But when I use the voice chat, the avatar first responds with an answer from the avatar's profile (Ann Therapist) before it actually recognizes the grounded reference. How can I avoid this? (I never want to use the default profile , I always it to answer from the grounded documents. )