floyo logo
Pricing
Create with Alibaba Happy Horse model now! Try here 👉
floyo logo
Pricing
Create with Alibaba Happy Horse model now! Try here 👉
Last updated
2026-04-14

This repository provides a set of nodes for ComfyUI that enable users to separate audio into distinct components, such as vocals, bass, and drums, as well as to manipulate and recombine audio tracks effectively. With capabilities including tempo matching and audio cropping, it serves as a powerful tool for audio processing and remixing.

  • Enables separation of audio into four distinct stems: vocals, bass, drums, and other instruments using Hybrid Demucs.
  • Provides functionality to combine multiple audio tracks through various mathematical operations like addition, subtraction, and averaging.
  • Includes tools for precise audio manipulation, such as tempo matching, cropping, and speed adjustments.

Context

This tool is an extension for ComfyUI that focuses on audio separation and manipulation, allowing users to isolate specific audio elements from tracks. Its primary purpose is to facilitate the editing and remixing of audio files by providing a user-friendly interface for audio processing tasks.

Key Features & Benefits

The nodes included in this repository offer practical features such as audio separation into four distinct stems, which allows for targeted editing of specific components. The ability to combine audio tracks using various mathematical methods enhances creative possibilities. Additionally, tools for tempo matching and audio cropping streamline the workflow for users looking to synchronize or trim audio segments.

Advanced Functionalities

The Audio Separation node utilizes Hybrid Demucs, a sophisticated model capable of accurately isolating different audio components. This model not only separates audio but also allows for further manipulation of the isolated tracks using the Audio Combine node, enabling users to create complex audio arrangements or remove unwanted elements.

Practical Benefits

By incorporating this tool into their workflows, users can greatly enhance their control over audio editing processes, leading to improved quality in their final outputs. The ability to precisely isolate and manipulate audio stems allows for greater creativity and efficiency in projects, ultimately saving time and effort in audio production.

Credits/Acknowledgments

The original authors and contributors of this repository are acknowledged for their work in developing the nodes and functionalities. The project is open-source, and the license details can be found within the repository.

Inner Nodes

AudioCombine
AudioCrop
AudioGetTempo
AudioSeparation
AudioSpeedShift
AudioTempoMatch
AudioVideoCombine