FILE // OTACON-AI-ECOSYSTEM TIER // OTACON CORE
Otacon Corefree · full source

Project 01 // Agent Platform

Otacon AI Ecosystem

The foundation of a Keep: persistent AI agents with their own identity, memory, and voice, routed through a provider-neutral core you run on your own hardware. See what a Keep is for where this is headed.

Right Now

Otacon Core today

Exactly what installing Otacon gives you the moment it finishes. No roadmap items included.

✓ Available the moment install finishes

  • ✓ Install Otacon locally on your own machine, one command, and it starts itself from then on
  • ✓ Default Model included: Ollama + a chat model sized to your GPU VRAM
  • ✓ Genome Voice Trainer included for custom Piper agent voices
  • ✓ Opens a local Command Center (Homelab-style deck) with Codec chat for Aria
  • ✓ Create and manage persistent agents
  • ✓ Maintain separate conversation history and long-term memory per agent
  • ✓ Configure agent identities and voice profiles
  • ✓ Provider-neutral capability routing: swap LLM/TTS/STT backends without touching agent identity
Normal Otacon install = Core + Default Model + Voice Trainer. Speech-to-text and image/video still need their own providers when you want those (see status).

Architecture

Built on a resource arbiter, not a script

User→ Agent (identity · memory · voice)→ Capability Router→ LLM / TTS / STT Providers

A provider-neutral resource arbiter sits beneath capability routing: workloads request logical CPU/RAM/GPU/VRAM and receive leases that respect placement, priority, and health, rather than agents hardcoding device indexes. Idle model residency can be reclaimed gracefully while active work is protected. Per-voice synthesis settings are configured individually rather than one global profile applied to every voice. This is the same architecture that, at full scale in the private Keep, extends to media, home, and multi-node GPU orchestration. See The Keep.

Install // Guided or manual

Install Otacon

Guided Windows Setup on the left. Manual step-by-step on the right: pick Windows, Ubuntu, or Advanced WSL first. Includes

Install Otacon

Guided Windows installer
Free Open source Runs locally No account No license key

Persistent AI agents, memory, a local Ollama chat model, Genome Voice Trainer, and the OtaconsKeep Command Center: on your hardware.

★ Recommended · Windows

Download one file. OtaconsKeep handles WSL, dependencies, Ollama, the default model, Voice Trainer, service setup, and health checks.

Download OtaconsKeep Setup
1. Download→ 2. Double-click→ 3. Approve Windows→ 4. OTACON IS READY

Typical first setup: ~10–30 minutes depending on your PC and internet. Windows may require one restart while enabling WSL: Setup resumes afterward.

What Setup handles

  • ✓ Enables / configures WSL2 when needed
  • ✓ Installs OtaconsKeep Core
  • ✓ Detects your hardware
  • ✓ Installs Ollama
  • ✓ Chooses a chat model based on available VRAM
  • ✓ Installs Genome Voice Trainer when supported
  • ✓ Runs installation health checks
  • ✓ Registers Otacon as a background service
  • ✓ Configures automatic startup after Windows login

What you need

  • Windows 10 or 11 (WSL2 support: Setup enables it if needed)
  • GPU: optional: CPU-only works
  • NVIDIA GPU: automatically detected when present
  • Internet: required for initial setup / model download
  • AI model: automatically sized for your hardware
  • Disk: leave ~10+ GB free for a comfortable first install

You’re finished when you see

✓ OTACON IS READY

Then open http://localhost:5757 to enter your Keep. Bookmark it: that address is local to this PC.

Having trouble? First-install help
Normal things you may see
Windows protected your PC / SmartScreen
  1. Click More info → Run anyway.
  2. Expected for a new unsigned Setup file.
Administrator / “allow this app” (UAC)
  1. Click Yes. Setup needs permission for WSL and services.
  2. A second Setup window may open: that is expected.
Windows asks you to restart
  1. Normal while enabling WSL. Save work and restart.
  2. After reboot, Setup should show WELCOME BACK, or double-click the same OtaconsKeep-Setup.bat again.
Ubuntu asks for a username / password
  1. Windows is creating OtaconsKeep’s local WSL Linux environment.
  2. Choose any username and a password you will remember (needed later for sudo).
“Updates are underway, please keep your computer on”
  1. Windows is finishing Linux/WSL features: not an Otacon crash.
  2. Can take 15–45+ minutes. Drive activity usually means it is still working.
Setup stays on one step for several minutes
  1. Dependencies and AI models can be large. Leave the window open.
Black terminal window with progress text
  1. That is Setup working. Do not close it until OTACON IS READY.
Something actually went wrong?
SETUP NEEDS HELP
  1. Recovery mode: not a silent crash. Press R to retry, O for logs, or X to exit.
localhost:5757 blank / broken
  1. Wait until Setup says OTACON IS READY, then open or refresh the address.
Installer file damaged / wrong encoding
  1. Delete the bad .bat.
  2. Re-download with the button above (or the GitHub ZIP fallback below): not browser Save As on a raw GitHub file.
'curl' is not recognized (manual path)
  1. Use the download button above instead, or try the PowerShell option under Manual → Windows.

Ask in Discord

Advanced · download notes, ZIP, Linux command

Recommended: use the download button above. It provides the Windows-formatted installer exactly as tested.

Prefer reviewing the code first? View source.

If the browser blocks the download, use the GitHub ZIP, extract it, then run OtaconsKeep-Setup.bat from the folder. Do not Save As on a single raw .bat from GitHub.

Already on Linux? Use the Manual column → Ubuntu / Debian, or paste:

curl -fsSL https://raw.githubusercontent.com/Otaconskeep/otacons-ai-ecosystem/main/install_otacon.sh | bash

Lighter install (skip chat model / voice trainer): put OTACON_INSTALL_DEFAULT_MODEL=0 and/or OTACON_INSTALL_VOICE_TRAINER=0 on the bash side of the pipe.

Method 2

Manual install: step by step

Baby steps. No Linux knowledge required. Pick your computer first: then only those steps appear.

What are you using?

1 Check your computer
  • ✓ Windows 10 or Windows 11
  • ✓ About 10+ GB free disk space
  • ✓ Internet connection
  • ✓ Permission to install software (Admin / Yes when Windows asks)

You do not need to install Python, Git, Ubuntu, Ollama, or WSL yourself. Setup handles that.

2 Open Command Prompt
  1. Press the Windows key.
  2. Type cmd.
  3. Click Command Prompt.

You should see a black window. That is correct.

3 Copy and paste
curl -L -o "%USERPROFILE%\Downloads\OtaconsKeep-Setup.bat" "https://otaconskeep.github.io/downloads/OtaconsKeep-Setup.bat" && start "" "%USERPROFILE%\Downloads\OtaconsKeep-Setup.bat"
  1. Click the black Command Prompt window.
  2. Right-click once to paste (or Ctrl+V).
  3. Press Enter.
CMD command didn’t work? Try PowerShell ▾
irm https://otaconskeep.github.io/downloads/OtaconsKeep-Setup.bat -OutFile "$env:USERPROFILE\Downloads\OtaconsKeep-Setup.bat"; Start-Process "$env:USERPROFILE\Downloads\OtaconsKeep-Setup.bat"
4 Approve Windows

If Windows says Windows protected your PC:

Choose More info → Run anyway.

If Windows asks Do you want to allow this app to make changes?

Click Yes.

5 Let Setup run
  • Leave the Setup window open.
  • First installation can take 15–45 minutes.
  • Some steps may appear to stop moving while Windows installs components or downloads AI models: that can be normal.
  • Your PC may restart once. Restart normally. After Windows starts again, Setup should continue. If it doesn’t, double-click OtaconsKeep-Setup.bat again.
6 You’re done

Finish line

✓ OTACON IS READY

When you see this message, installation is complete.

Open OtaconsKeep

Bookmark http://localhost:5757: this is your local OtaconsKeep.

1 Check your computer
  • ✓ Ubuntu or Debian (desktop or server)
  • ✓ ~10+ GB free · internet
  • ✓ Your user can use sudo (password when asked)
2 Open a terminal

Press Super/Windows key, type terminal, press Enter. Or use your existing SSH session.

3 Copy and paste
curl -fsSL https://raw.githubusercontent.com/Otaconskeep/otacons-ai-ecosystem/main/install_otacon.sh | bash
  1. Click the terminal window.
  2. Paste (right-click, or Ctrl+Shift+V).
  3. Press Enter.
  4. When asked for your password, type it (nothing appears) and press Enter.
4 Wait, then open Otacon

First run often takes 10–30 minutes. When it prints http://127.0.0.1:5757, open that address.

Finish line

✓ Install finished

Open http://127.0.0.1:5757

1 Only if Guided Setup won’t run

Most Windows users should use the Download button on the left. Use this path only when you need to set up WSL yourself.

2 Install Ubuntu (WSL)

Open PowerShell as Administrator, then run:

wsl --install -d Ubuntu

Restart if Windows asks. Open Ubuntu from the Start menu and create a username/password you’ll remember.

3 Run the Linux installer inside Ubuntu
curl -fsSL https://raw.githubusercontent.com/Otaconskeep/otacons-ai-ecosystem/main/install_otacon.sh | bash

When finished, open http://127.0.0.1:5757.

Need help? Click the message you see
Normal
Windows protected your PC
  1. More info → Run anyway.
Restart / WELCOME BACK
  1. Expected. Restart, then continue Setup.
Ubuntu username / password
  1. Windows creating the local Linux environment. Remember the password.
Actually stuck
'curl' is not recognized
  1. Use the left Download button, or PowerShell under Windows Step 3.
SETUP NEEDS HELP
  1. Press R to retry, O for logs.
sudo / Permission denied
  1. Use your Ubuntu password (invisible while typing).
localhost doesn’t load after READY
  1. Refresh, or ask in Discord.
ils>

✓ Verified, Windows installer, first install phase

◐ Not yet fully proven, post-reboot completion

The full "installed and usable after reboot" pass hasn't been independently confirmed end-to-end yet. This is a test-infrastructure limitation, not a known product defect: the automated Windows test lab runs Windows guests with nested virtualization disabled by default (a safety default to avoid a known guest-crash class), and Otacon's setup needs WSL2, which needs nested virt. The lab guest hung at the Windows boot screen after the WSL-enable reboot for that reason. The lab config has since been corrected (nested virt enabled) for future runs, but a corrected lab is not yet a green, repeatable release-gate result.

Release-gate criterion before this is called fully proven: the same download → hostile-path → double-click → WSL → reboot → resume → Core-ready → health-check sequence, green on real Windows 11 hardware or a VM with genuinely proven nested virtualization, not this lab alone.

✓ When it's done

Otacon is ready. Open it any time at:

Open Otacon →

Chat talks to Ollama at 127.0.0.1:11434. Voice Trainer status UI (when GPU path succeeded): http://127.0.0.1:8765.

Advanced: LAN access, repair, uninstall, and other options

Use Otacon from another device on your network

By default Otacon only answers on this machine. To reach it from your phone or another computer, find this PC's LAN IP (ip addr inside Ubuntu, or your router's device list) and visit http://<that-ip>:5757 instead.

Repair

Chat up but voice dead (Piper / TTS)? Do not only re-run Setup hoping it will fix itself, if the site answers, Setup can short-circuit to READY and skip TTS. Use explicit repair: OtaconsKeep-Setup.bat --repair (or choose [ P ] Repair when Setup offers Open / Repair / Reinstall). That restarts the otacon-tts unit, wake task, and spoken preview without wiping data. Full redo: --reinstall / --force. Status: OtaconsKeep-Setup.bat --status. Linux: otacon doctor.

Uninstall auto-start only

Run uninstall_otacon.bat (Windows). It removes the logon task and the systemd service only. Your Ubuntu environment, install folder, and data are untouched.

Optional environment variables (set before running install_otacon.sh)

VariableDefaultWhat it changes
OTACON_INSTALL_DIR~/otacon-ai-ecosystemWhere Otacon gets installed
OTACON_BUILD_NATIVE1Set to 0 to skip building the native .deb app
OTACON_INSTALL_DEB0Set to 1 to also install the built .deb automatically
OTACON_LAUNCH_WIZARD1Set to 0 to skip auto-launching the service at the end
OTACON_INSTALL_STT0Set to 1 to install optional Faster-Whisper speech-to-text
OTACON_RUN_TESTS1Set to 0 to skip the self-test suite (faster, less safe)
OTACON_INSTALL_VOICE_TRAINER1Set to 0 to skip Genome Voice Trainer
OTACON_INSTALL_DEFAULT_MODEL1Set to 0 to skip Ollama + default chat model
OTACON_LLM_MODEL(auto)Override the VRAM-picked Ollama model tag (e.g. qwen2.5:7b)
Requires sudo to install a handful of system packages (build tools, Python, WebKit libs) and to register the systemd service. The script explicitly refuses to run as the root user itself.
Default Model (included) installs Ollama with a VRAM-tier chat model: qwen2.5:1.5b (CPU/<6GB), 3b (≥6GB), 7b (≥8GB), 14b (≥16GB). Skip with OTACON_INSTALL_DEFAULT_MODEL=0 / OTACON_INSTALL_VOICE_TRAINER=0.

Full install documentation on GitHub →

Voice Training

Genome Voice Trainer, included with Otacon

Custom Piper voices for your agents. The normal Otacon install runs this for you. Standalone only if you skipped it or need to repair.

Standalone Voice Trainer (advanced)

Linux · Ubuntu / Debian / WSL

curl -fsSL https://raw.githubusercontent.com/Otaconskeep/otacon-voice-trainer/main/install_voice_trainer.sh | bash

Write-up: docs/WRITEUP.md

Status

What's real, what needs a provider, what's Keep-only

Straight from the repository's own status table. "Requires provider" means the routing and architecture exist and are testable today, you connect the backend. "Keep Reference" means it exists only in Antonio's private environment, not in this public repo at any maturity level yet.

Available now Requires provider Keep reference Planned

Persistent conversations

Available now

Long-term memory

Available now

Agent identity & voice profile config

Available now

Native installer & hardware planning

Available now

Backup / restore / safe uninstall

Available now

Agent chat (Default Model / Ollama)

Included with Otacon

Voice synthesis (Piper)

Requires provider

Genome Voice Trainer (GPU Piper)

Included with Otacon

Speech-to-text

Requires provider

Image / video generation

Requires provider

Single remote compute node

Requires provider

Agent rooms

Keep reference

Media automation, Home Assistant, search

Keep reference

Multi-node GPU orchestration

Keep reference

Discord / messaging integrations

Planned

Three Tiers

Otacon Core, the Keep Blueprint, and Otaconskeep Services

TierWhat it isStatusPrice
OtaconsKeep LiteFree · open source · self-hostedInstall todayFree
ExpansionMulti-agent layer; installer available today; GPU/creative sidecars optionalInstall today · fresh-PC notesFree
Keep BlueprintThe broader ecosystem architecture demonstrated by the real Keep, for example FOXDIE, automated media-library integrity with a reversible-only deletion pathReference onlyNot offered publicly
Otaconskeep ServicesArchitecture · deployment · supportAvailable nowContact for scope/pricing

There's no license key or artificial limit in Otacon Core. It's the real, complete public platform as it stands today. Otacon Core installs the foundation. The Keep Blueprint shows what that foundation can become.