Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > ComfyUI-Drama3 > Load Drama 3 Codec

ComfyUI Node: Load Drama 3 Codec

Class Name

Drama3CodecLoader

Category
audio/drama3
Author
PedroMarinhoDev (Account age: 50days)
Extension
ComfyUI-Drama3
Latest Updated
2026-11-12
Github Stars
0.17K

How to Install ComfyUI-Drama3

Install this extension via the ComfyUI Manager by searching for ComfyUI-Drama3
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-Drama3 in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

Load Drama 3 Codec Description

Loads the Fish S2-Pro codec for Drama 3 reference audio encoding and generated speech decoding.

Load Drama 3 Codec:

Drama3CodecLoader, displayed as “Load Drama 3 Codec” in the audio/drama3 category, loads the Fish S2-Pro audio codec used by Drama 3. A codec acts as a translator between ordinary sound and the compact numerical codes that the speech-generation system understands. This codec supports mono audio at 44.1 kHz and represents sound using 10 codebooks, with each frame covering 2,048 audio samples—approximately 21.5 frames per second. Loading it prepares the audio-conversion component needed for tasks such as encoding reference audio and decoding generated speech; the node does not generate speech or convert audio by itself. It uses the codec architecture supplied with the extension and loads compatible converted weights into it, while integrating with ComfyUI’s device and memory-management system. The codec is separate from the main Drama 3 generation model, and the extension is designed to use it before and after generation rather than keep it resident alongside the generation model during frame sampling.

Load Drama 3 Codec Input Parameters:

codec_name

Required dropdown selection identifying the codec weight file to load. The available options come from ComfyUI’s registered vae model locations, typically including ComfyUI/models/vae and any additional directories configured for that model category. Although this dropdown lists files from the VAE category, you must select weights compatible with the extension’s Fish S2-Pro codec architecture, not an ordinary image VAE or the main Drama 3 model. The node resolves your selection to a file path and loads its weights with strict compatibility checking, so choosing the correct file determines whether loading succeeds. The node description identifies the extension’s tools/convert.py codec conversion as the relevant preparation method. Options: the files currently returned by ComfyUI’s vae model listing; no fixed list of filenames is specified. Default: no explicit default is declared in the node schema. Minimum and maximum: not applicable because this is a file selection rather than a numeric setting.

Load Drama 3 Codec Output Parameters:

DRAMA3_CODEC

A loaded codec object with the custom connection type DRAMA3_CODEC. It contains the Fish S2-Pro codec model, including its audio encoder, codebook-based representation, and audio decoder, together with a ComfyUI model-management wrapper that handles its execution and offload devices. Think of this output as a ready-to-use audio translator rather than an audio recording: it contains neither generated speech nor saved speech codes. Connect it to compatible Drama 3 nodes that require a codec, such as the codec input of Drama 3 Decode, where it enables generated codes to be converted into 44.1 kHz audio. The loader constructs the model in float32 precision and uses ComfyUI’s selected execution and VAE offload devices; these settings are managed internally rather than exposed as controls on this node.

Load Drama 3 Codec Usage Tips:

  • Prepare the correct codec weights using the extension’s documented conversion workflow, identified in the node description as tools/convert.py codec. A file appearing in the dropdown does not guarantee that it is compatible.
  • Place the prepared codec file in a registered vae model directory, typically ComfyUI/models/vae. If a newly added file is missing from the dropdown, refresh ComfyUI’s model listing or restart ComfyUI, then check the selection again.
  • Keep the codec and the main Drama 3 model clearly distinguished in your filenames and workflow. Load Drama 3 Codec prepares the audio-conversion component; it does not replace the separate Load Drama 3 node used for the speech-generation model.
  • To decode a previously saved take, connect this node’s DRAMA3_CODEC output to Drama 3 Decode’s codec input, and connect Load Drama 3 Codes to its codes input. The resulting audio comes from the decode node, not this loader.
  • Use the same loaded codec output for compatible encoding and decoding steps when your workflow needs both. This node has only a file-selection control; it offers no sample-rate, precision, or quality settings to tune.
  • Address memory limits through ComfyUI’s device and memory-management configuration rather than looking for controls on this node. The loader uses float32 weights and ComfyUI’s VAE offload device, while the codec’s decoding implementation processes codes in windows to limit the memory needed for long takes.

Load Drama 3 Codec Common Errors and Solutions:

FileNotFoundError

  • Explanation: The selected filename can no longer be resolved to an existing file in ComfyUI’s registered vae locations. This can happen if you move, rename, or delete the file after selecting it. The exact exception text depends on ComfyUI.
  • Solution: Confirm that the codec file still exists in a registered vae directory, refresh the model listing or restart ComfyUI, and select the available file again. If you use an additional model directory, check that ComfyUI is configured to search it.

Error(s) in loading state_dict for Drama3Codec

  • Explanation: PyTorch could read the file, but its weights do not match the codec architecture expected by this extension. The loader uses strict checking, so incompatible weights are rejected rather than partially loaded.
  • Solution: Select the converted Fish S2-Pro codec weights intended for this extension. Do not select an image VAE, the main Drama 3 model, or an unconverted checkpoint with a different layout. If necessary, repeat the codec conversion using the extension’s documented instructions.

Missing key(s) in state_dict / Unexpected key(s) in state_dict / size mismatch for ...

  • Explanation: These are typical details accompanying a weight-loading compatibility error. Missing keys mean required model components were not found, unexpected keys mean the file contains components the loader does not expect, and size mismatches mean some stored weights have incompatible dimensions. The specific component names depend on the selected file.
  • Solution: Verify both the source checkpoint and the codec conversion method. Use weights prepared for the codec implementation in your installed extension, and check that your extension version and conversion instructions correspond to one another.

Checkpoint-reading or deserialization error

  • Explanation: ComfyUI’s model-file reader cannot interpret the selected file. The exact message varies with the file format and underlying library; possible causes include an incomplete download, a damaged file, or a file whose contents do not match its apparent model format.
  • Solution: Check that the download or conversion completed successfully, then obtain or regenerate a clean copy from a trusted source. Renaming a file’s extension does not convert its contents into compatible codec weights.

CUDA out of memory / torch.OutOfMemoryError

  • Explanation: Available device memory is insufficient for an allocation during model loading or later codec use. The loader constructs a float32 model and delegates device placement to ComfyUI, so the point of failure depends on your hardware, offload configuration, and other loaded models. An error during encoding or decoding may occur downstream rather than inside this loader.
  • Solution: Free memory by closing other GPU-heavy applications or unloading unused models, then retry. Review ComfyUI’s supported low-memory and offloading settings for your installation. There is no precision or memory-limit parameter on this node.

Value not in list: codec_name

  • Explanation: This is a possible ComfyUI validation message when a saved workflow references a codec filename that is not in the current dropdown options. It can occur after transferring a workflow to another computer or renaming a model file; the node itself does not define this error message.
  • Solution: Open the codec_name dropdown and choose a compatible codec file that exists on this installation. If no suitable file appears, add the prepared codec weights to a registered vae directory and refresh or restart ComfyUI.

Load Drama 3 Codec Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-Drama3
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

Load Drama 3 Codec