Aragon Maestro

490.00  ( 490.00  excl. tax /// 588.00  incl. tax)

Aragon Maestro: The Next Generation of AI-Driven Personal Assistants

Aragon Maestro is the core component of the Aragon Voice Control System, a comprehensive solution for smart home automation. The Maestro takes personal assistance to the next level with a significantly improved wake word engine, enhanced speech-to-text capabilities, Sound Event Detection and integrated on-board AI and To-do app.

 

Building on the success of the Aragon Master, the Maestro retains all its predecessor’s features, including on-board Node-RED and remote connection, while operating 100% offline with zero cloud dependencies.

 

Private by Design, the Maestro ensures that voice recognition runs entirely on the device – nothing is transmitted or recorded, providing users with complete control over their data.

SKU: M2 Category:

Description

The Aragon Maestro puts home automation control on a whole new level with the on-device addition of AI-driven personal assistance features. This represents an important advancement in voice control for home automation. We have built on the Aragon Maestro’s core smart appliance functionality such as controlling lights, shades, and heating, to produce a superior product with many new features that significantly enhance its overall capabilities.

 

Enhanced Wake Word Engine

The Maestro uses a rapid, extremely accurate wake word engine based on machine learning. This on-device technology allows it to control up to five satellites, making it ideal for common areas such as living rooms and kitchens. The high-quality analysis carried out locally reduces false detection, and performs optimally in all situations, including noisy environments. For quieter spaces like bedrooms, the original wake word system on the Satellite remains effective.

Advanced Hybrid Speech-to-Text System

The Maestro uses a hybrid speech-to-text system that combines Kaldi and Whisper. Kaldi quickly processes trained sentences, while Whisper understands an almost infinite range of speech. Whisper also excels over longer distances, and in noisy environments. Together, the two systems provide fast, accurate transcriptions, enhancing the user experience.

On-Board Large Language Model

By using an on-board large language model, the Maestro answers queries, such as “How to cook pasta” or “Convert 2 ounces to metric”, without needing Internet or cloud services. Information retrieval is quickly and reliably performed directly on-device.

Informational Commands:

Maestro increases the user’s knowledge base. Ask Maestro to:

– ‘Tell me, [something]’ – ask the on-board AI
– ‘Define [word]’ – to understand the meaning of a word
– ‘Explain [concept]’ – to learn something new
– ‘How to [task]’ – to receive help with a challenging job
– ‘Tell me, what is [topic]?’ – to get information on a specific subject
– ‘Convert [unit] to [unit]’ – to easily convert units of measurement
– ‘Calculate [math problem]’ – to quickly solve a calculation`

Important Note: The AI-powered Q&A feature utilizes the Mistral AI language model to provide voice-based responses. While this technology offers advanced language understanding and generation capabilities, it is essential to be aware that the AI may occasionally provide inaccurate or untruthful information.
Mistral AI is a French artificial intelligence startup. The company develops both commercial and open-source large language models, known for their transparency, portability, customisability, and cost-effectiveness.

 

Sound Event Detection

Sound Event Detection (SED). This is a sophisticated feature that enables real-time monitoring and classification of various acoustic events in the home environment. YAMNet neural network technology catagorises sounds into Alert, Human, Household, and Animal categories.

The SED monitoring interface is accessible via a Web browser interface. It offers a comprehensive view of detected sound events through five tabs: All Sounds, Alert, Human, Household, and Animal.

Each tab features:
– A time chart with color-coded dots representing detected sounds by category
– A Sound Event Gauge displaying instant detection values
– A Sound Activity time chart for temporal visualization
– A Sound Event list showing the latest events in text format

The ‘All Sounds’ tab provides an overview of all categories, while the other tabs offer detailed information specific to each sound type. This intuitive interface allows users to easily monitor and analyze various acoustic events in their environment, enhancing the overall functionality of the Aragon Maestro system.

 

Integrated To-Do and Reminder Application


The Maestro includes a built-in To-do and reminder application. In order to stay organized and on track with daily tasks users can update  lists with commands such as, “Put [task] on my to-do list” or “Remember to [task]”.

Tasks and reminders are added by voice and accessed from the onboard application. Users are able to easily mark tasks as completed, change content, and manage their to-do lists. The onboard application is simple to navigate and use via a Web browser interface.

For remote access, users can enable the Encrypted point-to-point remote connection (see Remote Connection below), allowing access of their to-do lists and reminders from anywhere. This ensures that they stay organized and productive on the go.

On-Device Text-to-Speech

The Aragon Maestro’s text-to-speech system runs entirely on-device, ensuring fast, secure voice generation. With a rich set of male and female voices, the Maestro provides a natural, engaging listening experience. Most spoken messages are generated in under a second, making interactions with the Maestro quick and seamless.

Advanced Customization with Whisper and Node-RED Integration

Thanks to the integration of Whisper speech-to-text technology, the Aragon Maestro offers unprecedented flexibility for power users and OEM customers. The system understands full English language input, which paves the way for custom voice command implementation beyond standard home automation tasks. Users can create customized voice commands not managed by the Maestro’s core functions, such as “Start charging car”, or “What is the level of the car battery?”. This capability, combined with Node-RED integration, significantly expands the Maestro’s potential applications, and allows for seamless connection to external services and devices.

Remote Connection with Tailscale

– Remote Connection to Your Smart Home with Aragon
– No Need to Open Firewall Ports
– Encrypted point-to-point connections using the open source WireGuard protocol
– Included in Aragon Maestro, No Subscription Required
– Technology: Tailscale VPN for Secure, Effortless Access

Read more: Remote Connection

Endless integration possibilities with Node-RED

Node-RED is an extremely powerful, easy-to-use visual programming tool. Tasks are automated by the user by interconnecting function blocks. No prior knowledge of a programming language is required to program Node-RED. Users have access to our large library of modules which can be copied and adjusted as needed.

 

Explore Aragon Node-RED: Example Code and Features

Smart Home Automation Made Easy: Introducing Node-RED Scheduling
Unlock powerful scheduling in Node-RED with this easy-to-use feature. Simply mark ‘ON’ and ‘OFF’ times in the calendar, enter corresponding voice commands, and automate your smart home devices – from lights and temperature to media control and more read more.

Visualize Your Data with LogView for proServ
Introducing LogView for proServ, a powerful feature that automatically logs and visualizes data points tagged with ‘#l’ in ETS. With LogView, you can easily track and analyze your data in customizable charts and dashboards, accessible from the Aragon menu read more.

Stay on Track: Aragon Voice Dialog’s Medication Reminder
Stay on top of your medication schedule with Aragon Voice Dialog’s Medication Reminder feature. This innovative tool sends timed voice prompts to remind you to take your medicine, and even responds to your voice commands to ensure you stay on track read more.

Code Smarter with Function-GPT: AI-Assisted Coding in Node-RED
Unlock the power of AI-assisted coding with Function-GPT, a Node-RED function node that integrates with ChatGPT to help you write code. Use it to generate code snippets, automate tasks, and debug your flows with ease, as demonstrated in this example flow that generates random numbers and handles different input types read more.

Node-Red gives endless possibilities for customization and personalisation.

System Overview

The Aragon Voice Control System is composed of:
– 1 Maestro and
– 1-10 Satellites available in two versions: WL (tabletop, RJ45/Wi-Fi) and PoE (flush mounted)

Key System Requirements

All Aragon devices must be installed on the same network. The Maestro is obligatory for system functionality, enabling voice command understanding, and interaction with the home automation system.

Configuration

The installation is designed to be truly plug-and-play.

At start-up, Aragon reads out the configured building structure, which includes:
– device names
– location
– type of device
– how to operate the device.

The received names will be trained, therefore understood by Aragon in the context, i.e. in complete sentences. Once trained, control and query of the installation is done by voice.

Aragon controls any IP-based Home Automation device, but for the following servers and gateways it is automatic, thanks to Plug-and-Play interaction/configuration:

– All KNX IoT compatible servers; e.g Schneider Wiser
– GIRA X1
– GIRA Homeserver with QuadClient (Version 4.10 or higher)
– THEBEN LUXORliving
– KNX-proServ
– JUNG Smart Visu Server
– Philips Hue Bridge
– PeaKNX YOUVI
– any KNX / IP-Interface (tunneling or routing), all gateways are supported, but will require manual configuration

Any other IP-based gateway with an open protocol can be used with Aragon, using a similar manual textual configuration.

Read more: User Guide

Hardware Spec

– SoC: Rockchip RK3588S
– CPU: Quad-core ARM Cortex-A76(up to 2.4GHz) and quad-core Cortex-A55 CPU (up to 1.8GHz)
– GPU: Mali-G610 MP4
– 64-bit 8GB LPDDR4X at 2133MHz
– Flash: 32GB
– Ethernet: one Native Gigabit Ethernet, and one PCIe 2.5G Ethernet
– USB: one USB 3.0 Type-A and one USB 2.0 Type-A
– LEDs: 4 x GPIO Controlled LED (SYS, WAN, LAN, LED1)
– Power supply: USB-C, support PD, 5V/9V/12V/20V input
– Ambient Operating Temperature: 0℃ to 70℃