(Note: This refers to avatar versions, not API versions.)
The Streaming API has officially deprecated Interactive Avatar Versions v1 and v2.
To continue using the service, please upgrade to v3 or higher immediately.
Deprecated versions will return errors. Full details & migration help.
Endpoint: /v1/streaming.new
Reference: https://docs.heygen.com/reference/new-session
Added new STT provider: AssemblyAI. It is now the default provider for English language transcription.
Endpoint: /v1/streaming.new
Reference: https://docs.heygen.com/reference/new-session
A new parameter activity_idle_timeout has been added to the streaming.new endpoint.
It specifies the maximum idle time (in seconds) after the last user interaction before the session is marked as inactive.Default: 120 seconds, Range: 30 to 3600 seconds. This gives you greater control over session lifecycle management and resource usage.
Endpoint: /v1/streaming.keep_alive
Reference: https://docs.heygen.com/reference/keep-alive
A new /v1/streaming.keep_alive endpoint has been introduced to reset the idle timeout counter for an active streaming session. Use this endpoint to keep the session alive and prevent it from timing out due to inactivity.
Endpoint: /v1/streaming/knowledge_base/*
Reference: https://docs.heygen.com/reference/list-knowledge-bases
We’ve added new Knowledge Base Management endpoints! You can now list, create, update, and delete knowledge bases for your interactive avatars.
Endpoint: /v2/avatar/{avatar_id}/details
Reference: https://docs.heygen.com/reference/retrieve-avatar-details
We’ve added the Retrieve Avatar Details endpoint! Retrieve detailed information for a specific avatar by providing its avatar_id.
Endpoint: /v2/voices/locales
Reference: https://docs.heygen.com/reference/list-all-locales-for-voices
We’ve added the List All Locales for Voices endpoint! Retrieve a full list of available voice locales (like en-US, pt-BR, etc.) that you can use with multilingual voices when generating videos. Before using, check if the voice supports locales via the support_locale attribute from /v2/voices.
Bring the full power of HeyGen's API to your favorite AI-enabled tools — like Cursor, Claude Desktop, and more — with zero manual implementation. Just install the HeyGen MCP Server and you're ready to go!
🔧 No need to write integration code
🧠 Works seamlessly with AI development environments
💡 Ideal for rapid prototyping and automation
Get started: https://docs.heygen.com/docs/heygen-mcp-server
Endpoint: /v2/streaming.list
Reference: https://docs.heygen.com/reference/list-sessions-v2
We have introduced a new v2 endpoint for listing sessions. This endpoint supports pagination and allows you to retrieve a list of all sessions with optional filtering by date range and status.
We've add support for choosing between Eleven Lab models such as eleven_flash_v2_5 or eleven_multilingual_v2. To see the trade offs either model, see: https://elevenlabs.io/docs/models
We've excited to partner with Gladia: a new speech to text provider, providing strong multi lingual support on speech recognition side.
More details here on how to configure these, see: https://docs.heygen.com/reference/new-session
