n8n-nodes-deepgram

This is an n8n community node package for interacting with the Deepgram API.

It currently includes the following nodes:

• Deepgram Transcriber: Transcribes pre-recorded audio files using the Deepgram API. Supports providing audio via URL or binary file input.
• Deepgram Speaker (TTS): Generates audio from text using the Deepgram Speak API (Text-to-Speech).

n8n is a fair-code licensed workflow automation platform.

Installation Operations Credentials Compatibility Usage Resources License

Installation

Follow the installation guide in the n8n community nodes documentation.

1. Go to Settings > Community Nodes.
2. Select Install.
3. Enter n8n-nodes-deepgram in the Enter package name field.
4. Agree to the risks of using community nodes: select I understand the risks, and I want to proceed.
5. Select Install.

After installing the node, you can use it like any other node. n8n displays the node in the node panel under Community > Installed.

Operations

• Deepgram Transcriber:
  • Transcribe audio from a public URL.
  • Transcribe audio from an n8n binary file property.
  • Supports various Deepgram models (including Nova 2 and Nova 3).
  • Allows specifying additional options like language, punctuation, diarization, smart formatting, keywords, and callback URL.
  • Optionally append metadata (endpoint, parameters, duration) to the output.
  • Choose between outputting the full raw transcript or just the transcript text.
• Deepgram Speaker (TTS):
  • Convert input text (up to 2000 characters) into spoken audio.
  • Select from various Deepgram Aura voice models.
  • Configure audio output options like encoding (MP3, WAV, etc.), container, sample rate, and bit rate.
  • Outputs the generated audio as an n8n binary file property.

Credentials

Requires Deepgram API credentials. The same credentials work for both the Transcriber and Speaker nodes.

1. Go to your Deepgram Console.
2. Navigate to API Keys.
3. Create a new API key or use an existing one.
4. In n8n, create new credentials for the Deepgram Transcriber node.
5. Enter your Deepgram API Key.
6. (Optional) If using a self-hosted or custom Deepgram endpoint, enter the Base URL.

Compatibility

Tested with n8n version 1.x.

Usage

1. Install the package in your n8n instance.
2. Add the Deepgram Transcriber node to your workflow.
3. Configure the node properties:
   • Select the Input Source Type (URL or Binary File).
   • Provide the Audio URL or Binary Property name.
   • Choose the desired Model.
   • Configure Additional Options as needed (language, punctuate, etc.).
   • Select the Output Format.
   • If using "Transcript Only" format, optionally specify a custom Transcript Field Name (defaults to transcript).
4. Connect the node and run your workflow.

Deepgram Speaker (TTS)

1. Install the package in your n8n instance.
2. Add the Deepgram Speaker (TTS) node to your workflow.
3. Configure the node properties:
   • Enter the Text to Speak.
   • Choose the desired Voice Model.
   • Configure Audio Options (encoding, container, sample rate, bit rate) as needed.
   • Specify the Output Binary Property name where the audio data will be stored (defaults to data).
   • (Optional) Provide an Output Filename (e.g., speech.mp3). If left empty, a default filename based on the output format will be used (e.g., deepgram_output.mp3).
4. Connect the node and run your workflow. The output item will contain the generated audio in the specified binary property, including the filename.

Resources

• n8n Community Nodes documentation
• Deepgram API Documentation
• GitHub Repository: Walid-Azur/n8n-nodes-deepgram

Consulting & Services

For custom node development, workflow automation consulting, or other n8n-related services, please contact:

Walid Boudabbous (Acceleate Consulting Estonia)
Email: [e-mail]

License

MIT