Introduction

This introduction opens a series of guides that aim to:

• Outline the benefits and use cases of self-hosting
• Describe the architecture, requirements, and needed assets for an installation
• Tell you how to configure your environment and set up your server for the installation
• Show you how to install the actual Deepgram application
• Provide a starting point for how to connect your own applications with your self-hosted Deepgram instance
• Help you plan your server maintenance and security practices

Why Self-Host

Using Deepgram as a service has a variety of benefits. First, it’s extremely fast to start developing with. Signing up, getting an API key, and getting your first voice AI response can take as little as a minute. Using Deepgram as a service also enables you to avoid all hardware, installation, configuration, backup, and maintenance-related costs.

With that being said, there are situations a self-hosted deployment might make sense. The most common use cases we have seen are when you have stringent performance or security requirements.

Performance

Certain use cases, like AI voicebots, have very sensitive latency and load requirements. If you need ultra-low latency with voice AI services colocated with your other services, self-hosting can meet these requirements.

Security

One of the common use-cases for self-hosting Deepgram is to satisfy security or data privacy requirements. In a typical self-hosted deployment, no audio, transcripts, or other identifying markers of the request content are sent to Deepgram. Your Deepgram components will only contact the Deepgram license server in order to validate the Deepgram components and models, as well as report usage information. The usage information reported for each request includes metadata such as audio duration or character count, features requested, and success response codes.

Furthermore, self-hosted deployments do not offer a method to persist request or response data on your own servers. Request and response data are not stored beyond the duration of the original API request.

Components

Before you deploy Deepgram, you’ll need to make effective design decisions about the components of your system, their relationships, and the interactions between components. Ideally, your architecture will meet your business needs, optimize both performance and security, and provide a strong technical foundation for future growth.

Deepgram provides a variety of components available for a self-hosted deployment. Many guides describe how to create a deployment using Deepgram’s required components, API and Engine. Some guides include details on additional components, such as the License Proxy in the diagram below. See Self-Hosted Add Ons for more details.