Category: StarterKit

The central component of the Starter Kit is the Raspberry Pi 5, which functions as the audio processor running the AudioworX Amp Application. The application receives raw audio input data from and sends back processed audio data to the Creative Sound Blaster X4, USB class-compliant audio interface (requires no drivers) with a stereo analog line-input port, and 4 stereo audio output ports for 7.1 audio playback.

The Starter Kit can be used with either passive or active speaker systems. The following sections show hardware connections in both passive and active speaker setups.

• Passive Speaker Setup with the JBL DSP-4086 Amplifier
• Active Speaker Setup without the JBL DSP-4086 Amplifier
• Verifying Hardware Connections with the AudioworX Starter Kit Utility

For the Starter Kit to be fully operational, the PC and the Raspberry Pi must be accessible to each other through a Local Area Network (LAN) and the preferred mode is via direct ethernet connection as indicated in the block diagram in Hardware Block Diagram – Passive Speaker Setup
However, in situations where a direct ethernet connection cannot be made, the Starter Kit can be configured to connect to the PC via Wi-Fi. For more details refer to “Set a Wi-Fi for Raspberry Pi to Connect” under SKUtility Configuration section in Starter Kit Utility Tool.

Passive Speaker Setup with the JBL DSP-4086 Amplifier

For a multi-channel passive speaker setup like in a car-cabin, an additional power amplifier is required since the Creative Sound Blaster X4 audio interface can only output line-level signal (+4 dBu, approx. 2.1Vrms), which is insufficient to drive passive speakers in most cases.

For such systems, the Starter Kit includes a JBL DSP-4086 car audio amplifier, a 6-input and 8-output amplifier that can drive 4Ω or 8Ω loudspeakers up to a maximum of 40W per channel. The on-board DSP in the amplifier is deliberately bypassed and the amplifier is configured to pass-through the 6 input audio channels to the first 6 output channels (the last 2 output channels are not used), with only power amplification to drive the speakers, to fully centralize all audio DSP to the AudioworX Amp application in the Raspberry Pi 5.

The below block diagram shows the hardware connections for such systems, indicating the cables required to interface the components of the Starter Kit.

1. Raspberry Pi Connections

At the Raspberry Pi end, the following connections need to be made:

1. Direct ethernet connection to the PC using an ethernet cable.
2. USB connection to the Creative Sound Blaster X4 using the USB type-A to USB type-C cable.
3. The power supply.

2. Creative Sound Blaster X4 Connections

At the Creative Sound Blaster X4 end, the following connections need to be made:

1. USB connection from the Raspberry Pi using the USB type-A to USB type-C cable (also mentioned in the Raspberry Pi connections).
2. Audio Outputs to the JBL DSP-4086 amplifier using 3.5 mm to RCA cables from the audio outputs ports labeled FRONT, SIDE and C/SUB.
3. Audio Input from an audio source such as a mobile phone or PC via the 3.5 mm audio input named LINE/OPT IN (on the same side as the audio outputs).

3. JBL DSP-4086 Connections

At the JBL DSP-4086 end, the following connections need to be made:

1. RCA Audio Inputs from the Creative Sound Blaster X4, using the 3.5mm to RCA stereo cables.
   1. FRONT output of the Sound Blaster X4 to Channels 1 and 2 of the JBL amplifier.
   2. C/SUB output of the Sound Blaster X4 to Channels 3 and 4 of the JBL amplifier.
   3. SIDE output of the Sound Blaster X4 to Channels 5 and 6 of the JBL amplifier.
2. Speaker Harness to the passive speakers via speaker cables (each speaker output has a + and – terminal labelled on the JBL speaker harness wires).
   1. Channels 1 and 2 to the Front Left and Front Right speakers.
   2. Channels 3 and 4 to the Center and Sub-woofer speakers.
   3. Channels 5 and 6 to the Side Left and Side Right speakers.
3. Power Supply from the 12V 20A SMPS Power Supply.
   1. Red wire to the +V terminal of the SMPS power supply.
   2. Black wire to the COM terminal of the SMPS power supply.

Ensure that the physical switches on the JBL amplifier are set to the following states:

1. “Input Level” switch to “Lo” (extreme left position),
2. “Turn-on Mode” switch to “Audio” (extreme right position).

In this state, the JBL amplifier will power on only when it receives audio inputs at its input channels 1 and 2. The below figure shows the states of the switches:

The following image shows the full set of connections (the SMPS power supply and passive speakers not shown) for a passive speaker setup.

This completes the connections between the components of the AudioworX Starter Kit in a passive speaker setup.

Active Speaker Setup without the JBL DSP-4086 Amplifier

For multi-channel active speaker setups in on test benches or listening rooms, where the speakers are powered, the audio output channels of the Sound Blaster X4 can be directly connected to the speakers, enabling an 8 speaker configuration compared to the 6 speaker configuration in passive speaker.

The below block diagram shows the hardware connections in an active speaker setup.

The connections at the Raspberry Pi end remain the same as in passive speaker setups (see figure Hardware Block Diagram – Passive Speaker Setup) and the connections at the Creative Sound Blaster X4 are mostly the same as in the passive speakers, with the following exceptions:

• There is an additional stereo output via the REAR output port to the physical Rear Left and Rear Right speakers.
• The RCA end of the output audio cables (3.5mm to RCA stereo cables) cables are connected to the speakers directly.

The below image shows the full set of connections (speakers not shown) in the active speaker configuration.

For simplicity, the rest of the documentation assumes that the connections between the Raspberry Pi 5 and the Creative Sound Blaster X4 are made as required, as they are the only essential connections for the proper functioning of the AudioworX Amp application.

The below section describes how the hardware connections can be verified using the Starter Kit Utility Tool.

Verifying Hardware Connections with the AudioworX Starter Kit Utility

Once the hardware setup is completed as described above and the software pre-requisite python libraries are installed on the PC, the connections can be verified using the diagnostics function of the AudioworX Starter Kit Utility Tool (explained in detail in SKUtility Diagnostics). The Starter Kit Utility diagnostics report identifies issues in the setup preventing proper functioning of the Starter Kit.

To validate the hardware connections, open a Command Prompt window in the “StarterKitUtility” directory of the GTT installation folder and execute the following command:

python3 SKUtility.py diag

This command generates a report as shown below. Refer to the Troubleshooting section for resolving issues that are detected by the diagnostics.

AudioworX Starter Kits received directly from Harman AudioworX contain hardware components that are pre-configured, and making hardware connections as described above are sufficient to get the Starter Kit ready for executing GTT workflows. If the components are purchased independently, write to Harman AudioworX support at AudioworXSupport@harman.com to receive detailed steps on configuring each component from ground-up.

Once the Starter Kit hardware has been set up and verified using the Starter Kit Utility diagnostics, it can interact with the Global Tuning Tool (GTT), a Windows Application for designing audio processing pipelines and signal flows, flashing them on target amplifiers (in our case, it is the Starter Kit) and tuning the Audio Objects (AO) in the signal flows in real-time.
GTT includes a sample project for the AudioworX Starter Kit that can be used a starting point for users to get familiar with the Starter Kit.

The example project for the AudioworX Starter Kit implements a simple signal flow, a custom tuning panel and some presets to showcase common AudioworX workflows that are explained in the Global Tuning Tool User Guide.

The following sections guide first-time users through the initial setup of AudioworX using the Starter Kit.

1. Importing The Example Project
2. Configuring AmpSrv2 Settings
3. Sending the Device Configuration to the AudioworX Starter Kit
4. Connecting to the Starter Kit for Tuning
5. Tuning the Audio Processing Pipeline
6. Saving and Recalling Tunings using Parameter Sets
7. Managing Presets using the Preset Controller

Importing The Example Project

The Home Screen of GTT provides options such as importing a project, creating a new project, open an existing project, explore some high-level features of GTT, etc.  For more details, refer Home screen options page.

To open the Starter Kit example project:

1. Go to the Sample Projects tab.
2. Double click the “AWXExampleProjectStarterKit”  project.
   
3. GTT may display a warning message. This warning is due to the project being similar to the default example project of GTT. Click “No” and proceed.

This opens the Device Designer window. The Device Designer is the first view of GTT that is shown on opening a GTT project. This tab shows a high-level view of the audio processing pipeline, showing the audio system/device and its abstractions as shown in the below figure.

• The Device level abstraction (AWXExampleProject1): This represents the actual physical device or the target device with physical audio inputs and outputs. The Starter Kit can have up to 2 audio input channels and 8 audio output channels which are provided by the Creative Sound Blaster X4 audio interfact. These channels are highlighted as “Physical Audio Inputs/Outputs” in the above image.
• The Physical core (arm64-v8a): This block represents the Digital Signal Processor (DSP) core that performs the audio processing. Although the Raspberry Pi 5 contains 4 64-bit Arm v8a Cores, currently, only one of the cores is supported for performing audio processing operations by the AudioworX Amp Application in the Starter Kit. Other AudioworX amplifiers may have more than one DSP core, each of which will be represented separately in the device designed view.
• The Virtual Core (Entertainment Task): This block represents a single virtual core performing a specific audio task. A virtual core processes audio at a user defined sample rate and block length, which can be configured in the “Edit Device” menu at the bottom of the “Device Designer” tab (also explained in Create Device File).
  In this example, we have one virtual core that takes a stereo audio input and gives out a 7.1 channel output (8 channels in total). This block is populated with core objects that can be dragged-and-dropped from the “Toolbox” pane on the left-side of the “Device Designer” view and interfaced together to perform operations such as sample rate conversion, buffering, routing, that may be required by the main audio processing instances called the “Xaf Instance”.
  For Core Objects Toolbox details, refer Core Objects Toolbox. A physical core may have more than one virtual core, each of which will be handled by a separate OS thread.
• The Audio Processing Instance (Xaf Instance): This block represents the core of the audio processing pipeline which is designed using the Signal flow Designer tool to implement a signal flow (Refer to Xaf Instance). 
  The rest of the Starter Kit documentation only briefly explains concepts that are relevant to a first-time user in the context of the AudioworX Starter Kit.

The Starter Kit is pre-configured with all the necessary data files to run the example project on first boot. The user may skip to the Connecting to the AudioworX Starter Kit for Tuning section for instructions on connecting to the Starter Kit. However, the following sections highlight some important steps that are specific to the Starter Kit for educational purposes before getting started with tuning of the signal flow itself.

Configuring AmpSrv2 Settings

The project’s AmpSrv2 settings must be configured for GTT to be able to communicate with the Starter Kit hardware for operations such as sending the audio processing signal flow and tuning.

Steps to configure AmpSrv2 settings:

1. On the Device Designer tab, click on AmpSrv Settings on the top ribbon bar and click Configure.
   
2. In the AmpSrv window, go to File > Options > Socket, and set the hostname to “raspberrypi.local” and the port to “25001”.
   
3. Click Ok to close the AmpSrv2 settings window and minimize the AmpSrv2 window.
4. On the Device Designer tab, click on AmpSrv Settings and click on Save from the drop-down menu to save the settings.

For AmpSrv2 Settings details, refer to the Connection page.

Sending the Device Configuration to the AudioworX Starter Kit

The device configuration contains information on how the physical and virtual cores, the Xaf instances (which contain the core of the signal processing flows) and other core AOs are interconnected to form the audio processing pipeline. As previously mentioned, the Starter Kit is pre-configured with the required data files, which includes the device configuration to run the example project on first boot.

For a different project or if any changes have been made to the example project, it is important to send the device configuration to the Starter Kit, to ensure the AudioworX Amp application can set up and execute the audio processing pipeline.

Steps to send the device configuration defined in the GTT project to the Starter Kit hardware:

1. On the Device Designer tab, click on Send Device Config.
   
2. Once the device configuration is sent successfully, if the new device configuration is different from the previously flashed device configuration on the Starter Kit, GTT may prompt the user to reboot the device.
   
3. In such cases, the AudioworX Amp application running in the Raspberry Pi 5 must be reset using the Starter Kit Utility tool by running the command:

python3 SKUtility.py rmt -rst

This command is explained under Reset AudioworX Amp Application in SKUtility Remote Control. For other AudioworX based amplifiers, a full reboot may be required.

Connecting to the AudioworX Starter Kit for Tuning

Once the device configuration has been sent to the Starter Kit hardware and the AudioworX Amp application has been reset (only required if prompted by GTT), the Starter Kit can be connected to from GTT for tuning.

Steps to connect GTT to the AudioworX Starter Kit:

1. Click the Connect Device button located in the top ribbon bar to initiate device connection.
   
2. A “Device Synchronization” window will pop up. Click on the Send button to send the tuning data.
   
3. Once the data has been sent and the device has been connected, the audio processing pipeline and its signal flow can be tuned in real-time.
4. To stop tuning and disconnect the Starter Kit from GTT, click the Disconnect Device button in the top ribbon bar of the Device Designer tab.
   

Tuning the Audio Processing Pipeline

The audio processing pipeline can be tuned in two ways: using the State Variables button in the top ribbon bar or by using a custom panel.

Tuning using State Variables

The State Variables explorer can be accessed from the top ribbon bar in the “Device Designer” tab as shown below.

The rows in the State Variable explorer such as “Gain_1_0_0”, “Limiter_1_0_0”, etc., correspond to the names of the specific AOs in the signal flow. Expanding the row items in the tree display the tunable parameters of the AOs can be set in the right-half of the window as show in the above figure.

Tuning using the Custom Panel

GTT provides the capability to design a custom User Interface specific to the signal flow in the project using drag-and-drop methodology, allowing the user to define their own tuning interface with simple steps from the “Panel Designer” tab. For details on custom panel creation and usage, refer Create a Custom Panel in GTT.

The example project includes a custom tuning panel that is pre-designed for tuning the audio processing pipeline by exposing essential tuning parameters for the AOs in the signal flow.

To launch the custom panel, go to the “Panel Designer” tab and double-click the item named “ExampleProject1” in the “Panels” pane on the left-hand side of the view as shown below.

This custom panel window has three tabs, each tab is tailored to facilitate tuning of audio parameters at different levels of detail.

• Main
• Mute/Gain/Inv
• SignalFlow

Main: The “Main” tab provides users with high-level volume and mute controls along with a global Bass and Treble control. The below image shows the “Main” tab of the Custom Panel.

Mute/Gain/Inv: The second tab is the “Mute/Gain/Inv” tab, which provides users with volume and mute controls for each of the 8 output channels in the audio processing pipeline, along with polarity controls for each speaker to minimize effects of phase destruction in the listening environment. The below image shows this tab of the Custom Panel.

SignalFlow: The “SignalFlow” tab, which provides more granular control of the overall signal flow, exposing controls such as individual speaker delays, EQ, limiter, noise gate, etc., which can be opened by clicking the individual blocks in the signal flow. The below image shows this tab of the Custom Panel.

Saving and Recalling Tunings using Parameter Sets

GTT allows users to save and recall tunings as presets from the “Parameter Sets” tab of GTT.

In the Starter Kit example project, 2 presets are created, namely “LFHighOnly” and “AllSpeakers”, where the first one mutes all speakers except for the Left-Front High speaker and the second preset enables audio outputs from all speakers.

For detailed information on the Parameter Sets tab and its features, refer to Parameter Set Overview.

Saving Custom Tunings

To save a tuning that you’ve made using the State Variables explorer or the custom panel, follow below steps:

1. Go to the Parameter Sets tab.
2. Click the New button on the top ribbon bar. This will create a new column named “Set3”.
3. Right-click the new column and select Retrieve to populate it with the current tuning settings.

To overwrite a preset, right-click the preset column and click Retrieve. Once the preset has been written, it is automatically saved with the project.

In the example project, the presets (described above) store all tuning parameters in the signal flow. However, they can be customized to store only a sub-set of all the tuning parameters in the signal flow using “Set Group” (left-side pane in the “Parameter Sets” view).

For details on creating a new set group with a sub-set of tuning parameters, refer to Create a Parameter Set in GTT.

Applying Tunings from a Saved Preset

A preset can be applied to the target by right clicking the preset name and clicking the Apply button as shown below:

Applying a preset sets the tuning parameters in a live signal flow running in a target device (here it is the Starter Kit), only if the device is connected. If the target device (here, the Starter Kit hardware) is not connected to the GTT session, only the local states of the parameters within GTT will be updated, which then needs to be sent while connecting the device as described in Connecting to the AudioworX Starter Kit for Tuning. 

Managing Presets using the Preset Controller

GTT provides a method for recalling and managing presets using data files stored in the target device’s flash memory using the “Preset Controller”, which is accessible from the “Device Designer” tab of GTT as shown below.

Presets made in the “Parameter Sets” tab can be mapped to slots in the “Preset Controller”, which represent preset files on the target device, for quick recall of presets.

In the Starter Kit example GTT project, there are 2 pre-made preset slots; “LFHighOnly” and “Allspeakers” (slots 1 and 2 respectively), which are mapped to the “LFHighOnly” and “AllSpeakers” presets in the “Parameter Sets” tab mentioned earlier.

The “Preset Controller” can be used to set presets by selecting the required Slot ID and clicking the “Load Slot” button at the bottom of the window. A default slot can also be selected to load on boot-up using the “Default Slot” drop-down menu. In the Starter Kit example project, the “AllSpeakers” preset is made as default.

For details on configuring the Preset Controller, refer to Configuring Preset Controller.

This section outlines the process to be followed for creating a new project with a custom device type, providing instructions for efficiently creating a project environment that supports custom audio configurations and leveraging the full capabilities of GTT with the Starter Kit.

1. Creating a Project
2. Adding a Device to the Project
3. Leveraging AudioworX Audio Features

Creating a Project

Steps to create a new empty GTT project:

1. Click the + button in the “Projects tab” of the “Home Screen” of GTT.
2. Select Empty Template and provide project title and description.
3. Click the Create button to create new empty project.
   

Adding a Device to the Project

The first step after creating a GTT project is to add a device to the project, either from a template or by defining a custom device, within which the audio processing pipeline will be implemented. GTT provides the capability for defining a device with a custom number of audio inputs, outputs, virtual cores processing audio data at custom sample rates and block lengths, etc., to build audio systems requiring precise audio data handling, ensuring that the audio system is fully customized to support the project’s specific audio processing needs.

GTT can also learn the capabilities of an AudioworX amplifier within which the audio processing pipeline is to be built by querying the target amplifier.

The following section provides instructions how to define a custom device that is compatible with the Starter Kit hardware.

Configuring a Custom Device Compatible with the AudioworX Starter Kit

A custom device can be defined by creating a device file that contains information on the physical cores available on the target amplifier, virtual cores configured for audio processing at custom sample rates and block lengths within each physical core, and groups of audio input and output channels to stream audio data to the virtual cores.

The device file can be customized to support complex audio workflows, enabling scalability, optimized resource usage, and the flexibility to handle different audio streams simultaneously.

Steps to create a custom device file:

1. Click the + button in the “Devices” pane of the “Device Designer” tab.
2. Select the Create Device File option on the right side of the window.

   

This will open a “Device File Editor” where a user can define the high-level capabilities of the device.

First, a physical core must be created to describe the DSP core that the target amplifier is equipped with and provide details on its MIPS and core-type. An AudioworX supported amplifier hardware may have more than one DSP core that can be used for audio processing, each of which need to be separately added to the device file.

Steps to create the physical core in the device file editor for the Starter Kit:

1. Right-click Physical cores and click Add Physical Core.
   
2. Select Physical Core 0 choose the arm64-v8a option in the Core-type drop-down menu, which corresponds to the processor core type of the Raspberry Pi 5. This is an essential step to ensure compatibility of the GTT project with the Starter Kit.
   

The Starter Kit currently only supports one Physical Core of type “arm64-v8a” despite the Raspberry Pi 5 having a quad-core processor.

Further steps on adding and defining “Virtual Cores” and detailed information on the “Device File Editor” can be found in Create Device File.

Device level Input and Output Groups can also be edited in the “Device File Editor” to suit the needs of the audio processing pipeline. Here, the user is expected to set the properties of audio input and output data streams that the physical device is capable of, including the number of channels, sample rate of the audio data and the format of the audio data.

The Starter Kit currently only supports the following configuration:

1. Sample rate: 48,000 Hz
2. Max. number of input channels: 2
3. Max. number of output channels: 8
4. Audio data format: Float

As a last step after editing the device file, click Save Device Template and save the device.flash file on local file system. This device.flash file must be sent to the Starter Kit hardware using the Starter Kit Utility to make the kit fully aware of the exact device defined in GTT.

For details on how to send the device.flash file refer to “Flash a Device File” in the “SKUtility Configuration” section of AudioworX Starter Kit Utility Tool – SKUtility.

Discovering the Device Properties from the AudioworX Starter Kit Hardware

GTT can learn the device capabilities using the “Discover Device” option in the “Add Device” menu. The discovered device type can then be used in the GTT project to design the audio processing pipeline.

Ensure that the AudioworX Amp application is running in the Raspberry Pi for GTT to be able to learn from the Starter Kit Hardware. Refer to the “SKUtility Diagnostics” section of AudioworX Starter Kit Utility Tool – SKUtility.

Steps to query the Starter Kit hardware to discover the device properties based on the device.flash file that has been already flashed on to the kit:

1. Launch AmpSrv2 from the Desktop or from the Windows Start bar.
2. Go to File > Options > Socket and set the host name as “raspberrypi.local” and port as “25001”.
3. While keeping the AmpSrv2 window open, in the “Add Device” window of GTT, click the Discover Device button. This will get the device information from the Starter Kit hardware and display as a new device under the templates list (“Device-4” in the below figure).
4. Select the device and click Ok to add the device into the GTT project.
   

   

For the device capabilities to be discoverable from the Starter Kit hardware, the Starter Kit should have been previously flashed with a device.flash file.
By default, the Starter Kit is configured with a device.flash file that corresponds to the device type used in the example project. This file can be restored in case it was deleted using the command.
python3 SKUtility.py conf -rstAll
More details, refer to “Reset All Configurations to Default” under “SKUtility Configuration“.

Leveraging AudioworX Audio Features

GTT facilitates enhanced testing and experimentation with custom audio setups, providing flexibility and control over the audio workflow.

The GTT system includes “Xaf instances” and Core AOs using which developers can enhance audio processing systems, ensuring seamless integration and efficient handling of audio tasks. AudioworX provides a robust framework for manipulating audio data while taking full advantage of GTT’s advanced functionalities to deliver optimized performance and enhanced audio output.

The center pane of the “Device Designer” tab in GTT is the “Device View” where the audio processing pipeline is built. Refer to the “Importing The Example Project” for a detailed explanation of the device level abstractions in the Device View.

The “Toolbox” pane on the left-side provides a set of objects that can be dragged, dropped, and interconnected to complete the audio processing pipeline.

The following steps show how a simple audio processing pipeline can be developed:

1. Connect the pins in physical core (block named “Device-4”) and virtual core (block named “Virtual Core 0”).
2. Drag and drop an “Xaf Instance” from the left-side “Toolbox” pane to “Virtual Core 0” and set input and output channels as 8 and 8 (illustrative only), respectively.
3. Now, connect the pins on the added “Xaf Instance” and “Virtual Core 0”.
   
4. Double-click the Xaf Instance to edit/add the AOs and define a signal flow.
5. Drag and drop the required AOs into the designer and connect them as required. The below figure shows a Gain object of 8 channels that is added to the signal flow.
   
6. Once the connections have been made as required, click Save to save the signal flow design and click Go Back to return to the “Device view”.
7. The audio processing pipeline or the signal flow can now be sent to the Starter Kit as described in the Sending the Device Configuration to the AudioworX Starter Kit and can be tuned there after as described in the Connecting to the Starter Kit for Tuning.

The AudioworX Starter Kit is a complete hardware setup that provides a comprehensive experience of AudioworX (AWX) and its features with minimal setup time. The Starter Kit, much like other Harman amplifiers based on AudioworX, depends on the Global Tuning Tool (GTT) Windows application for designing signal flows, sending them to the target amplifier, tuning in real-time, selecting presets, etc. The basic idea of the Starter Kit is to provide a smooth onboarding experience into AudioworX workflows by simplifying steps for integration into an audio system using commonly available hardware components and supplementary software tools.

This document offers comprehensive guidance on configuring the AudioworX Starter Kit as a target amplifier. It includes detailed instructions, connection diagrams, software installation, GTT integration for signal flow design and tuning, and troubleshooting tips. For in-depth information on GTT features like parameter sets, master preset controller, and panel design, refer to the AudioworX documentation.

Related Topics

• Starter Kit Requirements
• Hardware Components Setup
• Getting Started with the Example Project
• Creating a New Project on GTT

The AudioworX Starter Kit consists of hardware and software components that need to be set up for functioning as a target amplifier which can interface with GTT to provide a comprehensive experience of AudioworX. The following sections cover the hardware and software components required and provide steps to install software pre-requisites.

• Hardware Components
• Required Software Applications

Hardware Components

Following is a list of hardware components required to setup the Starter Kit:

• Raspberry Pi 5 8 GB RAM variant (low-cost pocket-sized Linux computer).
• 27 W power adapter for the Raspberry Pi (USB Type-C).
• Micro SD Card with custom 64-bit Raspbian OS image (plugged into the Raspberry Pi 5).
• Sound Blaster X4 USB Audio Interface for multichannel analog audio input and output.
• JBL DSP 4086 power amplifier with 6 input channels (for passive speaker setups only).
• 12V 20A SMPS power supply for the amplifier (for passive speaker setups only).
• Ethernet cable for connecting the Raspberry Pi 5 to the PC (recommended)
• Speakers for audio output (6 passive speakers or 8 active speakers).
• A stereo analog audio source (PC or mobile phone with audio out).
• Windows PC

The JBL DSP-4086 amplifier and the 12V 20A power supply are only required if the speakers being used for audio playback are passive. For active speaker setups (speakers with their own power), the Creative Sound Blaster X4 can be directly connected without the JBL amplifier. Refer to Hardware Components Setup for more details.

Required Software Applications

Following is a list of software applications to be installed on a Windows PC to operate the Starter Kit:

• Global Tuning Tool Pre-Requisites
• Global Tuning Tool v24
• Python 3.12

Download and install the above-mentioned pre-requisites on a Windows 10 or 11 PC. For steps on how to install GTT and its pre-requisites, refer to open Installation and Licensing guide.

The GTT installation includes the Starter Kit Utility tool, a Python command-line application that provides the ability to remotely operate the Starter Kit hardware over the network, for actions like verifying hardware connections, audio device configuration, resetting the AWX Amp application running on the Raspberry Pi, rebooting the Raspberry Pi, etc. For more details, refer to Starter Kit Utility Tool.

This tool is installed in the following directory (default GTT installation folder):

C:\Program Files\Harman\HarmanAudioworX\StarterKit\StarterKitUtility

To install the pre-requisite python packages:

1. Go to the “StarterKitUtility” directory, type “cmd” in the address bar, followed by the enter key to open a Command Prompt window.
2. Run the following command to install the python pre-requisites as shown below.

python3 -m pip install -r requirements.txt