Microsoft AutoGen Voice Integration

AutoGen by Microsoft enables conversational multi-agent systems. Adding voice makes these conversations audible.

What is AutoGen?

AutoGen is a framework for building multi-agent conversational AI:

• Agents chat with each other to solve problems
• Supports human-in-the-loop
• Code execution capabilities
• Function calling

Adding Voice to AutoGen

from autogen import AssistantAgent, UserProxyAgent
from langvoice_sdk.tools.autogen_tools import LangVoiceAutoGenToolkit

# Initialize voice toolkit
toolkit = LangVoiceAutoGenToolkit(api_key="your-langvoice-key")

# LLM configuration with voice functions
llm_config = {
    "config_list": [{"model": "gpt-4o", "api_key": "your-openai-key"}],
    "functions": toolkit.get_function_schemas(),
}

# Create voice-enabled assistant
assistant = AssistantAgent(
    name="voice_assistant",
    system_message="You can generate speech using LangVoice tools.",
    llm_config=llm_config,
)

# Create user proxy
user_proxy = UserProxyAgent(
    name="user",
    human_input_mode="NEVER",
)

# Register voice functions
for func in toolkit.get_functions():
    user_proxy.register_function(function_map={func.__name__: func})

# Start conversation
user_proxy.initiate_chat(
    assistant,
    message="Summarize today's AI news and speak the summary aloud"
)

Multi-Agent Voice Conversations

Create agents that speak to each other:

host = AssistantAgent(name="host", llm_config=llm_config)
guest = AssistantAgent(name="guest", llm_config=llm_config)

# Each can generate audio with different voices

Use Cases

• Automated customer service with audible responses
• Educational tutoring systems
• Multi-character storytelling
• Technical support agents

Build conversational voice agents with AutoGen and LangVoice!