> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# EmptyLatentAudio - ComfyUI Built-in Node Documentation

> Complete documentation for the EmptyLatentAudio node in ComfyUI. Learn its inputs, outputs, parameters and usage.

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/EmptyLatentAudio/en.md)

The EmptyLatentAudio node creates an empty latent tensor for audio processing. It generates a blank audio latent representation with a specified duration and batch size, which can be used as a starting point for audio generation or processing workflows. The node automatically calculates the appropriate latent dimensions based on the audio duration and sample rate.

## Inputs

| Parameter    | Data Type | Required | Range        | Description                                           |
| ------------ | --------- | -------- | ------------ | ----------------------------------------------------- |
| `seconds`    | FLOAT     | Yes      | 1.0 - 1000.0 | The duration of the audio in seconds (default: 47.6)  |
| `batch_size` | INT       | Yes      | 1 - 4096     | The number of latent images in the batch (default: 1) |

## Outputs

| Output Name | Data Type | Description                                                                                                                                                                                                                                                                                                                        |
| ----------- | --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `LATENT`    | LATENT    | Returns an empty latent tensor for audio processing with the specified duration and batch size. The tensor has a shape of \[batch\_size, 64, length], where length is calculated from the audio duration and sample rate. The output also includes metadata indicating the type is "audio" and a temporal downscale ratio of 2048. |

***

**Source fingerprint (SHA-256):** `02d7623358a6cc8200e74ab36f6a83bcfa059abd6ef91e3fd1c3732706926033`