ElevenLabs Speech to Speech icon

ElevenLabs Speech to Speech

Converte áudio de uma voz para outra usando a API da ElevenLabs

Converter Voz

AI-generated

Overview

This node converts audio from one voice to another using the ElevenLabs API. It is useful for scenarios where you want to transform spoken content recorded in one voice into a different voice, such as creating voiceovers, dubbing, or personalized audio messages. For example, you could convert a podcast episode recorded by one speaker into the voice of a famous narrator or generate audio content in a preferred voice style.

Properties

Name Meaning
Fonte do Áudio Source of the audio file to process. Options: "Arquivo Binário" (binary file uploaded to the node) or "URL" (link to an audio file).
Campo Binário Name of the binary property containing the audio file when using a binary file as source. Default is "data".
URL do Áudio URL of the audio file to process when using URL as source.
ID da Voz The voice ID to which the input audio will be converted. This identifies the target voice style in the ElevenLabs system.
Nome do Campo Binário de Saída Name of the binary property where the processed audio will be stored. Default is "data".
Opções Avançadas Collection of advanced options:
- Aumento de Similaridade Number between 0 and 1 controlling how closely the resulting voice resembles the original voice. Higher values mean more similarity. Default is 0.75.
- Estabilidade Number between 0 and 1 controlling the stability and variability of the voice. Higher values mean more stable and less variable voice. Default is 0.5.
- Formato de Saída Output audio format. Options include MP3, FLAC, WAV, and various PCM formats with different sample rates. Default is MP3 44.1kHz 128kbps.
- Idioma de Origem Language of the input audio. Options include automatic detection and many languages like English, Spanish, Portuguese, French, German, etc. Default is automatic ("auto").
- Idioma de Destino Language of the output audio. Can be the same as the source language or any supported language. Default is "same" (same as source).

Output

The node outputs an array of items, each containing:

  • json: An object indicating success (success: true), the voice ID used, and the advanced options applied.
  • binary: Contains the processed audio file under the user-defined binary property name. The audio is encoded in the selected output format (e.g., MP3, FLAC, WAV).
  • pairedItem: References the original input item index.

If the input was a binary file, the node reads the audio data from the specified binary field; if it was a URL, it uses that URL directly. The output binary data contains the transformed audio in the chosen voice and format.

Dependencies

  • Requires an API key credential for the ElevenLabs API.
  • Uses the ElevenLabs speech-to-speech endpoint.
  • No additional external dependencies beyond the ElevenLabs API and n8n's built-in HTTP request and binary data helpers.

Troubleshooting

  • No binary data found: If the node is set to use a binary file but the specified binary property does not exist or is empty, it will throw an error. Ensure the input contains the correct binary field.
  • Invalid voice ID: Providing an incorrect or unsupported voice ID may cause API errors. Verify the voice ID is valid in your ElevenLabs account.
  • Unsupported audio format: The node supports specific output formats; selecting an unsupported format might cause issues.
  • API errors: Network issues or invalid API credentials will result in errors. Check your API key and network connectivity.
  • Language mismatch: Setting incompatible source and target languages might affect conversion quality.

Links and References

Discussion