OpenRouter TTSInstall

Summary

Convert the required text to speech through the configured OpenRouter API using the selected model and voice. Return the audio as binary data and include the model, voice, format, and filename in the output JSON.

Inputs

• Model (required) — TTS model ID, chosen from the loaded model list or entered with an expression.
• Text (required) — Text to convert to speech; supports n8n expressions.
• Voice (required) — Voice identifier. The field suggests OpenAI voice names or an ElevenLabs voice ID or name; the selected model determines provider support.
• Output Format — Choose MP3 or raw PCM. Defaults to MP3.
• Speed — Playback speed from 0.25 to 4; sent only when different from 1, and supported by some providers.
• Binary Property Name — Name of the output binary property; defaults to data.

Output shape

One output item per input item, each with JSON metadata and binary audio data.

JSON includes model, voice, format, and fileName (for example, speech.mp3). Audio is stored under the configured binary property name, with MIME type audio/mpeg for MP3 and audio/pcm otherwise. If Continue On Fail is enabled, an item error is returned in JSON instead of audio.

Examples

Example 1: Generate a spoken version of text

Choose a TTS model, provide the text, and set a provider-supported voice. Optionally set the output format, speed, and binary property name.

Links

• n8n community nodes installation guide