mirror of
https://github.com/supermemoryai/supermemory.git
synced 2026-09-30 01:51:28 +00:00
5b2a435e37
examples/voice_agent.py runs in console mode or joins LiveKit rooms. It scopes memory from dispatch metadata, SUPERMEMORY_CONTAINER_TAG, or the participant, preloads the caller's profile, and greets returning callers by name. Tested on LiveKit Cloud: a second call greeted the caller by name and used a fact from the first call. The example is not included in the wheel or sdist.
71 lines
2.6 KiB
Python
71 lines
2.6 KiB
Python
"""Voice agent that remembers each caller across calls.
|
|
|
|
Run from this folder after filling in .env (see README.md):
|
|
|
|
python voice_agent.py console # talk through your mic in the terminal
|
|
python voice_agent.py dev # join LiveKit rooms, e.g. from the Agents Playground
|
|
"""
|
|
|
|
import json
|
|
import logging
|
|
import os
|
|
|
|
from dotenv import load_dotenv
|
|
from livekit import agents
|
|
from livekit.agents import AgentServer, AgentSession, ChatContext, JobContext
|
|
from supermemory_livekit import SupermemoryAgent, SupermemoryLiveKit
|
|
|
|
load_dotenv()
|
|
logger = logging.getLogger("voice-agent")
|
|
|
|
INSTRUCTIONS = (
|
|
"You are a friendly voice assistant. You remember this caller across calls. "
|
|
"Use what you know naturally, and do not mention the memory system. "
|
|
"When the caller asks you to remember something, call remember. Keep replies short."
|
|
)
|
|
|
|
server = AgentServer()
|
|
|
|
|
|
# An agent_name turns on explicit dispatch, which is how job metadata reaches the agent.
|
|
# Leave it empty to join every new room, which the Agents Playground needs.
|
|
@server.rtc_session(agent_name=os.getenv("LIVEKIT_AGENT_NAME", ""))
|
|
async def entrypoint(ctx: JobContext):
|
|
memory = SupermemoryLiveKit(api_key=os.environ["SUPERMEMORY_API_KEY"])
|
|
|
|
# Your backend can pass the caller id in dispatch metadata. SUPERMEMORY_CONTAINER_TAG
|
|
# pins every call to one caller while testing. Console mode has no remote participant.
|
|
metadata = json.loads(ctx.job.metadata or "{}")
|
|
container_tag = metadata.get("container_tag") or os.getenv("SUPERMEMORY_CONTAINER_TAG")
|
|
if not container_tag and ctx.is_fake_job():
|
|
container_tag = "console_user"
|
|
|
|
await ctx.connect()
|
|
if container_tag:
|
|
memory.bind(container_tag=container_tag, session_id=ctx.room.name)
|
|
else:
|
|
participant = await ctx.wait_for_participant()
|
|
memory.bind(participant=participant, session_id=ctx.room.name)
|
|
logger.info("memory scoped to container tag %s", memory.container_tag)
|
|
|
|
# The caller's profile goes into the first turn, so the greeting can use it.
|
|
chat_ctx = ChatContext()
|
|
await memory.preload(chat_ctx)
|
|
|
|
session = AgentSession(
|
|
stt="deepgram/nova-3:en",
|
|
llm="openai/gpt-4.1-mini",
|
|
tts="cartesia/sonic-3",
|
|
)
|
|
memory.attach(session)
|
|
await session.start(
|
|
room=ctx.room,
|
|
agent=SupermemoryAgent(memory, chat_ctx=chat_ctx, instructions=INSTRUCTIONS),
|
|
)
|
|
await session.generate_reply(
|
|
instructions="Greet the caller. If you already know them, welcome them back by name."
|
|
)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
agents.cli.run_app(server)
|