Skip to main content

Deployment from AWS marketplace

This guide will walk you through deploying the Phonexia Speech Platform 4 as a virtual appliance on Amazon Web Services (AWS).

What you'll need:

  • The Licensed models ZIP file (provided by Phonexia)
Preparation: Verify your AWS account setup

Before starting the deployment, make sure you have everything ready:

AWS Account and Permissions​

You need an active AWS account with permissions to:

  • Launch and configure EC2 instances
  • Create and manage VPC (Virtual Private Cloud) resources
  • Allocate Elastic Block Store (EBS) volumes for storage
  • (Optional) Set up CloudWatch for monitoring your instance

If you're unsure about your permissions, contact your AWS administrator.

SSH Key Pair​

You'll need an SSH key pair to securely connect to your instance after deployment.

  • If you already have one, note its name—you'll select it during setup
  • If you need to create one, you can do this during the EC2 launch process or beforehand in the AWS EC2 console under Network & Security > Key Pairs
Step 1: Launch Virtual Appliance

Subscribe to Phonexia Speech Platform​

First, you need to subscribe to the Phonexia Speech Platform product in the AWS Marketplace:

  1. Log in to the AWS Management Console
  2. Navigate to AWS Marketplace (search for it in the top search bar)
  3. Click on Discover products in the left sidebar
  4. Search for "Phonexia Speech Platform"
  5. Select the product from the search results
  6. Click View purchase options
  7. Review the pricing and terms, then click Subscribe
  8. Wait a moment for the subscription to be processed (this usually takes less than a minute)

Once you see "Subscription successful," you're ready to launch your instance.

Launch and Configure Your EC2 Instance​

Now you'll create the actual virtual machine that runs the Speech Platform:

  1. Go to your marketplace subscriptions
  2. Find the Phonexia Speech Platform product and click Launch
  3. Choose the Launch method based on your preference
Launch from EC2 console

If you choose the Launch from EC2 console method instead of the One-click option, please refer to the official AWS documentation for detailed instructions, as this approach is beyond the scope of this guide.

  1. For One-click configure the following settings:

    Region:

    • Select in which region should be deployed

    Instance Type:

    • For most use cases, select g4dn.2xlarge (this includes a GPU for optimal performance)
    • If you plan to process large volumes or need more power, choose a larger instance type

    VPC and Subnet:

    • Use your default VPC or select a custom one if you have specific networking requirements

    Security Group:

    • Create a new security group or select an existing one
    • Important: Make sure it allows:
      • SSH (port 22) from your IP address for console access
      • HTTP (port 80) and/or HTTPS (port 443) from your IP address or CIDR range for web interface access

    Key Pair:

    • Select the SSH key pair you prepared earlier (or create a new one if needed)
Need Help with EC2?

For information about how to set up an AWS EC2 instance, please refer to the official AWS documentation

Choosing the Right Instance Size

When choosing an EC2 instance type, select the one most suited for the technologies you wish to run.

Preferred instance type is set to g4dn.2xlarge, which is considered baseline for running Speech Platform. When you run into service being slow, you should consider redeploying and selecting an instance type with more resources.

  1. Review your settings and click Launch

Wait for Instance to Start​

After clicking Launch, your instance will begin initializing:

  1. Go to the EC2 Dashboard in AWS Console
  2. Click on Instances in the left sidebar
  3. Find your Phonexia Speech Platform instance in the list
  4. Wait until the Instance State shows Running (this usually takes 2-3 minutes)
  5. Once running, find and copy the Public IPv4 address or Public IPv4 DNS—you'll need this to access the platform
Platform Boot Time

Even after the instance status shows Running, the Speech Platform itself needs a few more minutes to fully start up. If you can't access the web interface right away, wait 3-5 minutes and try again.

Step 2: Upload licensed models and configure the platform

The virtual appliance distribution contains only the bare Speech Platform system. To actually breathe life into the system it's necessary to upload the licensed models to the appliance and configure the technologies.

These steps can be done either semi-automatically via the platform GUI, or manually via SSH access to the platform console.

The Virtual Appliance contains a Filebrowser tool, which simplifies some management tasks like uploading the models and accessing and editing the appliance configuration files.
The Filebrowser is accessible at http://<virtual appliance IP address>/filebrowser.

Open the Filebrowser and click on the upload icon in the top right corner, then click on the "File" option, browse for the licensed models ZIP file (licensed-models.zip). A small pop-up window in the bottom right corner will show the upload progress. You can also click on the up arrow to show the graphical progress bar.

When the upload is finished (progress is at 100 %), Filebrowser automatically starts extracting the uploaded ZIP file and then runs the automatic configuration script. While these processes are executed, the upload progress pop-up stays shown with the progress stuck at 100 %, seemingly doing nothing - don't worry, this is a sign of the background tasks still in progress.

When the progress pop-up disappears, it means that the extraction and configuration is finished and the system is ready for use.

The automatic configuration detects which models and licenses were extracted from the ZIP file and automatically turns on the corresponding technologies and configures them to use the provided models, and also detects the hardware — if a GPU is present, it enables GPU processing.