Demeterics Speech GenInstall

Summary

Generate speech audio from input text using selectable text-to-speech (TTS) providers via the Demeterics API, returning the result as audio binary, a signed URL, or both.

Inputs

• Provider (required) — Choose the TTS provider (Groq Orpheus, OpenAI TTS, ElevenLabs, Google Cloud TTS, Murf.ai).
• Model (required) — Select the TTS model specific to the chosen provider.
• Voice (required) — Select the voice for speech generation, options vary by provider.
• Text (required) — Text string to convert to speech (max characters depend on provider: OpenAI 4096, ElevenLabs 5000, Google 5000, Murf 10000).
• Output Format (required) — Choose the audio output format supported by the selected provider (e.g., MP3, WAV, Opus).
• Speed — Set playback speed from 0.25 to 4 (only for providers supporting speed: OpenAI, Google, Murf).
• Voice Instructions — Optional instructions to style voice tone (only for OpenAI GPT-4o Mini TTS model).
• Language — Language code (ISO 639-1), used for speech synthesis language (only supported by Murf.ai).
• Output (required) — Determine output delivery as binary audio data, signed URL, or both; the URL expires in 15 minutes.

Output shape

A list of items matching input, each with JSON metadata about the generated speech (id, provider, model, voice, duration, cost, usage, metadata), plus audio data returned as binary, URL, or both depending on settings.

Supports multiple input items processed independently; audio binary is base64-decoded and prepared for download or further processing; errors for each item are returned in JSON when continuing on failure is enabled.

Examples

Example 1: Convert a short text to speech using OpenAI TTS, selecting voice and model, getting MP3 audio binary for saving.

Set Provider to 'OpenAI TTS', choose Model and Voice, enter Text, select MP3 format, leave Speed at default if not needed, choose Output as 'Binary'.

Links

• n8n community nodes installation guide