Discourse Extended icon

Discourse Extended

Read and manage Discourse AI artifacts, categories, posts, topics, personal messages, uploads, and Data Explorer queries

Actions25

Post → Generate TTS Audio + Spaces Upload

AI-generated

Summary

Synthesizes TTS audio from text using ElevenLabs and optionally uploads artifacts to DigitalOcean Spaces.

Inputs

  • cleanedText (required) — Cleaned TTS text to synthesize (for example from Prepare for TTS output).
  • voiceId (required) — ElevenLabs voice ID.
  • modelId (required) — ElevenLabs model ID (default: eleven_multilingual_v2).
  • audioFormat — Final output format. WAV uses PCM generation for deterministic stitching (default: mp3).
  • generateTranscript — Whether to request alignment/timestamps and build transcript artifacts (default: true).
  • transcriptOutput — Which transcript artifacts to create (default: textAndAlignmentJson).
  • chunkStrategy — How to split text before sending chunks to ElevenLabs (default: paragraphAware).
  • includeBinaryData — Whether to return generated audio/transcript files as n8n binary output (default: true).
  • audioBinaryFieldName (required) — Output field name for audio binary data (required if includeBinaryData is true).
  • transcriptBinaryFieldName (required) — Base name used for transcript binaries (required if includeBinaryData and generateTranscript are true).
  • spacesOptions — DigitalOcean Spaces configuration for uploading artifacts (bucket, folder path, public visibility).

Output shape

One output item per input text chunk containing JSON with audio/transcript metadata and optionally binary data.

If text is split into chunks, the node emits one item per chunk. Binary fields are named according to configuration (e.g., 'audio' for audio, 'transcriptText' or 'transcriptAlignment' for transcripts).

Examples

Example 1: Synthesize a short post into MP3 and include the transcript.

Set cleanedText to the post body, voiceId to a specific ID, audioFormat to mp3, generateTranscript to true, and includeBinaryData to true.

Example 2: Generate WAV audio for a long post and upload to Spaces.

Set audioFormat to wav, includeBinaryData to true, audioBinaryFieldName to 'audio', and configure spacesOptions with bucket and folderPath.

Links

Discussion