Convert Text to Speech Using OpenAI TTS API

Overview

This workflow converts any text into natural-sounding speech using OpenAI's Text-to-Speech (TTS) API. It is ideal for generating audio content for accessibility, voiceovers, podcasts, or language learning. The workflow is triggered manually for testing, but can easily be adapted to run on a schedule, via webhook, or as part of a larger automation.

Node-by-Node Breakdown

• When clicking "Test workflow" (Manual Trigger) — No auth. This node starts the workflow when you click the "Execute Workflow" button in the n8n editor. It has no parameters and is used for testing.
• Set input text and TTS voice (Set) — No auth. This node defines the text to be converted and the voice to use. It outputs a JSON object with two fields:
  • input_text: The text you want to turn into speech (e.g., "The quick brown fox jumped over the lazy dog.")
  • voice: The voice model to use (e.g., "alloy"). Other options include "echo", "fable", "onyx", "nova", and "shimmer".
• Send HTTP Request to OpenAI's TTS Endpoint (HTTP Request) — API Key auth (via OpenAI credential). This node sends a POST request to https://api.openai.com/v1/audio/speech with the following body parameters:
  • model: tts-1 (the TTS model)
  • input: The text from the previous node (using expression {{ $json.input_text }})
  • voice: The voice from the previous node (using expression {{ $json.voice }}) The request is authenticated using an OpenAI API key stored in an n8n credential. The response is a binary MP3 audio file.

Setup Instructions

1. OpenAI Account: You need an OpenAI account with API access. Obtain an API key from the OpenAI dashboard.
2. n8n Credential: In n8n, create a new credential of type "OpenAI". Paste your API key into the credential.
3. Configure the HTTP Request Node: The node is already set to use the OpenAI credential. Ensure the credential is selected in the node's settings.
4. Customize Text and Voice: Modify the input_text and voice values in the Set node as needed. The maximum input length is 4,000 tokens.

Use Cases and Variations

• Automated Podcast Generation: Replace the manual trigger with a Schedule Trigger to generate daily audio summaries from a news feed.
• Accessibility: Use a Webhook Trigger to accept text from a form or chat, then return the audio file to the user.
• Language Learning: Combine with a Translate node to convert foreign language text into speech.
• Content Creation: Integrate with a Google Drive node to save the generated MP3 files automatically.
• Voice Customization: Experiment with different voice options (alloy, echo, fable, etc.) to match the tone of your content.

{
  "id": "6Yzmlp5xF6oHo1VW",
  "meta": {
    "instanceId": "173f55e6572798fa42ea9c5c92623a3c3308080d3fcd2bd784d26d855b1ce820"
  },
  "name": "Text to Speech (OpenAI)",
  "tags": [],
  "nodes": [
    {
      "id": "938fedbd-e34c-40af-af2f-b9c669e1a6e9",
      "name": "When clicking \"Test workflow\"",
      "type": "n8n-nodes-base.manualTrigger",
      "position": [
        380,
        380
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "1d59db5d-8fe6-4292-a221-a0d0194c6e0c",
      "name": "Set input text and TTS voice",
      "type": "n8n-nodes-base.set",
      "position": [
        760,
        380
      ],
      "parameters": {
        "mode": "raw",
        "options": {},
        "jsonOutput": "{\n  \"input_text\": \"The quick brown fox jumped over the lazy dog.\",\n  \"voice\": \"alloy\"\n}\n"
      },
      "typeVersion": 3.2
    },
    {
      "id": "9d54de1d-59b7-4c1f-9e88-13572da5292c",
      "name": "Send HTTP Request to OpenAI's TTS Endpoint",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        1120,
        380
      ],
      "parameters": {
        "url": "https://api.openai.com/v1/audio/speech",
        "method": "POST",
        "options": {},
        "sendBody": true,
        "sendHeaders": true,
        "authentication": "predefinedCredentialType",
        "bodyParameters": {
// ... truncated (copy to see full JSON)