Skip to content

Use a Persona per Device

Alpha channel only

The prerelease channel carries this content. A stable or testing install does not have it, and a plain pip install does not get it: the install needs --pre. Read Release Channels before you mix channels.

This page gives each device its own persona on one OVOS install. You have more than one device, for example a speaker in the kitchen and one in the bedroom. You want a different persona in each room.

What the setup includes

  • A summon command, such as "talk to ada", sets persona_id on the session of the device that heard it.
  • Other devices keep their own persona, or none.
  • A release command, such as "stop talking to ada", clears persona_id on that session only.

Steps

Community servers are best-effort demos

The public OVOS servers exist for easy onboarding and demos only. They are best-effort, not optimized, carry no uptime guarantees, and may vanish at any time. OVOS will be slow and unreliable if you rely on them. The official recommendation is to self-host — or skip servers entirely: fully offline plugins exist for everything.

  1. Install the packages:

    pip install --pre ovos-persona ovos-openai-plugin

  2. Save two personas, personas/ada.json and personas/bo.json. This is ada.json:

    {
  "name": "ada",
  "handlers": ["ovos-chat-openai-plugin"],
  "ovos-chat-openai-plugin": {
    "api_url": "https://llama.smartgic.io/v1",
    "key": "sk-none",
    "model": "qwen2.5:7b",
    "system_prompt": "Your name is Ada. You answer in at most fifteen words. When asked your name, say Ada."
  }
}

    For bo.json, change the name to bo and change the system prompt to the name Bo.

  3. Save this script as devices.py and run it from the directory that holds personas. It sends the voice commands of two devices through the persona intent matcher:

    from ovos_bus_client.message import Message
from ovos_bus_client.session import Session
from ovos_persona import PersonaService
from ovos_utils.fakebus import FakeBus

svc = PersonaService(bus=FakeBus(), config={"personas_path": "personas"})


def heard(text, sess):
    msg = Message("recognizer_loop:utterance", {"utterances": [text]}, {"session": sess.serialize()})
    match = svc.match_high([text], "en-US", msg)
    return match.updated_session if match else sess


kitchen = heard("talk to ada", Session("kitchen"))
bedroom = heard("talk to bo", Session("bedroom"))
for sess in (kitchen, bedroom, Session("hall")):
    print(sess.session_id, "->", sess.persona_id)
for sess in (kitchen, bedroom):
    print(sess.session_id, "says:", " ".join(svc.query("What is your name? Answer with your name only.",
                                                     sess.persona_id, sess)))
kitchen = heard("stop talking to ada", kitchen)
print("kitchen after release ->", kitchen.persona_id)

Output of a run:

kitchen -> ada
bedroom -> bo
hall -> None
kitchen says: Ada.
bedroom says: Bo.
kitchen after release -> None

The persona is in the session, not in the service. A device that sends its session with each message, such as a HiveMind satellite, keeps its persona until a release command. A client can also set persona_id in the session that it sends, with no voice command.

Troubleshooting

Problem Cause Fix
All devices change persona together The devices share one session id Give each device its own session id
The summon does nothing The name does not match a loaded persona Use the name from the persona file. Look in the log for Failed to load