
Connectors / AI · HTTP Connector
Automate Deepgram with Tray.ai
Convert text to speech, create an agent configuration and update a project in Deepgram, and start a workflow when Deepgram sends “Transcription completed”.
What can you do with Deepgram on Tray.ai?
Deepgram is a voice AI platform that turns recorded or live audio into text, generates speech from text, analyzes text, and runs voice agents. Tray.ai connects to Deepgram through the HTTP Connector, so your workflows can call the Deepgram API alongside the rest of your stack.
- When Deepgram sends “Transcription completed”, attach the transcript to the call record in your CRM.
- Transcribe pre-recorded audio from support calls and add the text to the ticket in your help desk.
- Get project usage each month and send the totals to finance for cost tracking.
- Connects through
- HTTP Connector
- Authentication
- An API key in a request header
- Base URL
- https://api.deepgram.com/v1
Set it up
Connect Deepgram to Tray.ai
Each step uses Deepgram's own API details, from its documentation.
Step 1
Save Deepgram's API key
Save the key as a Tray.ai authentication. The HTTP Client puts it in the request header, read from $.auth. The key stays out of the workflow and out of the logs.
Step 2
Point the HTTP Client at the API
Add an HTTP Client step and use https://api.deepgram.com/v1 as the base URL. Or paste a cURL example from Deepgram's API docs, and Tray.ai fills in the method, URL, and headers.
Step 3
Start with “List agent configurations”
It's one of 19 documented actions, under Voice agents. The response comes back as data the next step can map into your other apps.
Step 4
Start workflows from Deepgram events
Give Deepgram the URL of a Tray.ai webhook trigger. “Transcription completed” can then start a workflow.
Give your AI agents Deepgram tools
Build an MCP server on Deepgram in Agent Gateway and choose which actions your agents can call. Every call runs with access control and an audit trail. How Agent Gateway works
- Transcribe pre-recorded audio
- Convert text to speech
- Create an agent configuration
- List projects
- List project keys
- Get project usage
- Transcribe live audio
- Stream text to speech
What the Deepgram API covers
6 resources and 19 documented actions, each one a call the HTTP Connector can make from a Tray.ai workflow. Connecting it takes a few clicks: here's how.
Speech to text
- Transcribe pre-recorded audio
- Transcribe live audio
Text to speech
- Convert text to speech
- Stream text to speech
Voice agents
- Create an agent configuration
- List agent configurations
- Update agent metadata
- Delete an agent configuration
Projects
- List projects
- Get a project
- Update a project
- Delete a project
API keys
- List project keys
- Create a project key
- Delete a project key
Usage and billing
- Get project usage
- Get usage breakdown
- Get project balances
- List project requests
Events Deepgram can send to a Tray.ai webhook trigger
- Transcription completed
Checked against Deepgram's own documentation on 2026-10-22
How Tray.ai makes this work
Deepgram plugs into the whole Tray.ai platform
Intelligent Integration
Integrate and automate across 10,100+ apps, databases, APIs, and services with visual workflows, error handling, and observability.
Learn more →Agent Builder
Build AI agents that read, write, and take action in Deepgram — with guardrails, audit, and human-in-the-loop.
Learn more →Agent Gateway for MCP
Expose Deepgram actions as governed MCP tools — observable, rate-limited, authenticated.
Learn more →Tray Helix
Build apps on Deepgram data with your AI assistant, then deploy, run, and govern them on Tray.ai.
Learn more →More ai apps on Tray.ai
Dedicated connectors and other apps Tray.ai reaches in the same category. See every ai connector →
Frequently asked questions
What can a Tray.ai workflow do in Deepgram?
+
The Deepgram API documents 19 actions across 6 resources (Speech to text, Text to speech, Voice agents, Projects, API keys, Usage and billing). For example: Transcribe pre-recorded audio; Convert text to speech; Create an agent configuration. A workflow can make any documented call through the HTTP Connector.
How does Tray.ai authenticate with Deepgram?
+
The Deepgram API uses an API key in a request header. You save it once as a Tray.ai authentication, and the HTTP Connector adds it to every request.
Which Deepgram events can start a Tray.ai workflow?
+
Deepgram sends one event to a webhook URL, for example: Transcription completed. A Tray.ai webhook trigger receives them and starts a workflow.
See Deepgram working against your stack.
We'll walk through a tailored demo with your systems plugged in.
Thanks — someone from our team will be in touch soon.