ElevenLabsSpeechToSpeech - ComfyUI Built-in Node Documentation - ComfyUI

The ElevenLabs Speech to Speech node transforms an input audio file from one voice to another. It uses the ElevenLabs API to convert speech while preserving the original content and emotional tone of the audio.

Inputs

Common Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use for speech-to-speech transformation. The selected model determines the available voice settings listed below.	DYNAMIC_COMBO	Yes	`eleven_multilingual_sts_v2` `eleven_english_sts_v2`
`voice`	Target voice for the transformation. Connect from Voice Selector or Instant Voice Clone.	CUSTOM	Yes	-
`audio`	Source audio to transform.	AUDIO	Yes	-
`stability`	Voice stability. Lower values give broader emotional range, higher values produce more consistent but potentially monotonous speech (default: 0.5).	FLOAT	Yes	0.0 - 1.0
`output_format`	Audio output format (default: “mp3_44100_192”).	COMBO	Yes	`"mp3_44100_192"` `"opus_48000_192"`
`seed`	Seed for reproducibility (default: 0).	INT	Yes	0 - 4294967295
`remove_background_noise`	Remove background noise from input audio using audio isolation (default: False).	BOOLEAN	Yes	-

eleven_multilingual_sts_v2 and eleven_english_sts_v2 Inputs

Both models provide the same set of voice settings below.

Parameter	Description	Data Type	Required	Range
`speed`	Speech speed. 1.0 is normal, <1.0 slower, >1.0 faster (default: 1.0).	FLOAT	Yes	0.7 - 1.3
`similarity_boost`	Similarity boost. Higher values make the voice more similar to the original (default: 0.75).	FLOAT	Yes	0.0 - 1.0
`use_speaker_boost`	Boost similarity to the original speaker voice (default: False).	BOOLEAN	Yes	-
`style`	Style exaggeration. Higher values increase stylistic expression but may reduce stability (default: 0.0).	FLOAT	Yes	0.0 - 0.2

Outputs

Output Name	Description	Data Type
`audio`	The transformed audio file in the specified output format.	AUDIO

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): a3cd602181d134b9ab517bfac092ea30b62ef5a9942a905c0c3e6959b34370ca