audexum

n8n community node for Audexum TTS API. Convert text to speech with 43 voices and 33 languages.

Package Information

Released: 4 mo ago
Downloads: 4 weekly / 281 monthly
Latest Version: 0.1.2
Author: Petar

Documentation

n8n-nodes-audexum

An n8n community node for the Audexum text-to-speech API.

Convert text to speech with 43 voices across 33 languages directly inside your n8n workflows.

Installation

In your n8n instance: Settings → Community Nodes → Install → n8n-nodes-audexum

Or via CLI:

npm install n8n-nodes-audexum

Setup

  1. Sign up at audexum.com — free tier is 10,000 characters/month, no credit card required
  2. Copy your API key from the dashboard
  3. In n8n, create a new Audexum API credential and paste your key

Usage

Add an Audexum TTS node to your workflow:

Parameter Description
Text The text to synthesize (up to 5,000 characters)
Voice Select from 43 voices (e.g. af_heart, am_michael)

The node outputs a binary WAV file — connect it to a Write Binary File node, an email attachment, or any downstream node that accepts binary data.

Example workflow

HTTP Request → Audexum TTS → Write Binary File

or

Schedule Trigger → Read Database → Audexum TTS → Send Email (with attachment)

Pricing

Plan Price Characters
Free €0 10K/mo
Starter €4/mo 100K/mo
Pro €12/mo 500K/mo
Scale €30/mo 2M/mo
PAYG €3 1M credits

Full pricing at audexum.com/pricing

API docs

audexum.com/docs

License

MIT

Discussion