Explain the Gemini Live API streaming capabilities in Google ADK for real-time voice/video.

Streaming & Gemini Live API in Google ADK

Google ADK integrates with the Gemini Live API for low-latency, bidirectional voice and video streaming, enabling real-time conversational AI.

---

Streaming Architecture

---

Capabilities

---

Audio Streaming Agent

pythonCopy
from google.adk.agents import Agent
from google.adk.runners import Runner, RunConfig

agent = Agent(
    name="voice_assistant",
    model="gemini-2.5-flash",
    instruction="You are a voice assistant. Respond naturally and conversationally.",
    tools=[get_weather, search_web],
)

# Enable streaming
run_config = RunConfig(
    streaming_mode="SSE",
    response_modalities=["AUDIO"],
    output_audio_transcription=True,
    speech_config={
        "voice_config": {
            "prebuilt_voice_config": {
                "voice_name": "Charon"
            }
        }
    }
)

---

Running Streaming in Web UI

bashCopy
# Launch with streaming support
adk web my_agent
# Click the microphone icon to start voice streaming

---

Text Streaming (SSE)

pythonCopy
from google.adk.runners import Runner, RunConfig

config = RunConfig(streaming_mode="SSE")

async for event in runner.run_async(
    user_id="user-1",
    session_id=session.id,
    new_message="Explain quantum computing",
    run_config=config,
):
    # Events arrive as they're generated
    if event.content and event.content.parts:
        for part in event.content.parts:
            if part.text:
                print(part.text, end="", flush=True)

---

Streaming with Tools

---

Streaming vs Non-Streaming

Learn more at Streaming.