EmptyYuE2LatentAudio - ComfyUI Built-in Node Documentation - ComfyUI

This node creates an empty audio latent for YuE2, sized to a chosen duration and batch count. It produces silent placeholder audio data that later nodes can fill in during audio generation.

Inputs

Parameter	Description	Data Type	Required	Range
`seconds`	Length of the audio latent to create, in seconds (default: 120.0). The number of latent frames is calculated from this value, with a minimum of 1 frame.	FLOAT	Yes	0.04 to 1000.0 (step 0.04)
`batch_size`	Number of audio latents to create in one batch (default: 1).	INT	Yes	1 to 4096

Outputs

Output Name	Description	Data Type
`LATENT`	An empty audio latent containing a tensor of zeros, sized by `batch_size` and the frame count derived from `seconds`. It is tagged as audio type with a temporal downscale ratio of 1920.	LATENT

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): 3397e3feb534c87d9790bbfe36e8e641476d9b6aa00d75a4e6d0c32f4a948563