<< Click to Display Table of Contents >> Navigation:  Nodes > Tasks > AI Transcription > Elevenlabs text to Speech

 

Purpose

 

This node performs AI text to speech conversion using ElevenLabs cloud service.

 

Before using this node you need to have an account in their website, with a valid plan to use the service. Then get an apiKey from your account and place it in the config -> apiKey property

 

Note: The Model and Voice properties can not be selected until a valid apiKey has been added.

 

An easy way to configure the authentication credentials is to store them in the Workflows -> Workflows Credentials Manager, which permits to store credentials and re-utilize them later by simply selecting their name in the node contextual menu (right click on the node) -> Credentials list.

 

 

Inputs

 

txt: A plain text file (.txt) encoded in UTF-8. One audio file will be generated per input text file.

 

Outputs

 

output: Audio file received after AI generation.

 

Standard properties

Object Name

Color

code

dstPath

 

Node specific properties

 

Dst. Path: Output path where the AI generated audio will be stored.

 

Model: The ElevenLabs AI model to use for text to speech synthesis. The list of available models is loaded dynamically from the API once a valid API key is provided.

 

Voice: The voice to use for reading the input text. The list of available voices is loaded dynamically from your ElevenLabs account once a valid API key is provided.

 

Format: The output audio format and quality. Available options include various MP3 bitrates (e.g. mp3_44100_128) and PCM sample rates (e.g. pcm_44100).

 

Stability: Controls how consistent the generated voice sounds (range: 0.0 to 1.0, default: 0.50). Higher values produce a more stable, predictable delivery; lower values allow more expressive variation between takes.

 

Similarity Boost: Adjusts how closely the output matches the selected target voice (range: 0.0 to 1.0, default: 0.80). Higher values increase similarity but may introduce artifacts if set too high.

 

Style: Amplifies the speaking style of the selected voice (range: 0.0 to 1.0, default: 0.00). Higher values exaggerate the voice's natural style. Increasing this setting may reduce overall stability.

 

Use Speaker Boost: When enabled, applies an additional processing pass to boost voice clarity and similarity to the original speaker. This may increase latency.

 

Config -> API key: Place your API key to access the service.

 

 

NOTE: For troubleshooting, execute your workflow and keep an eye in the Render Log panel, which will show any messages received from the Elevenlabs service.

 

Developer reference (Python - ElevenLabsTextToSpeech.[Property ID], API Rest - node/ElevenLabsTextToSpeech/property/[Property ID])