{"slug": "separate-dialogue-music-and-sound-effects-in-comfyui", "title": "Separate Dialogue, Music, and Sound Effects in ComfyUI", "summary": "A developer released ComfyUI-DBS, an unofficial ComfyUI integration of Dynamic Band Splitting (DBS) that separates a mixed video soundtrack into dialogue, music, and sound effects stems so users can strip dialogue while keeping music and effects. The custom node runs locally on CPU or NVIDIA CUDA GPU, requires ComfyUI v0.38.0+ and Python 3.10+, and ships with workflows for rebuilding a video from music plus SFX or saving speech separately; the two models total about 23 MB and are downloaded with pinned revisions and SHA-256 verification. The developer notes the stems are estimates, so dialogue can bleed into other tracks and short excerpts should be checked by ear first.", "body_md": "A video arrives with one mixed soundtrack. You want to replace the dialogue, but keep the background music and sound effects. Muting the audio removes all three.\n\nI built [ComfyUI-DBS](https://github.com/hiroki-abe-58/ComfyUI-DBS) to bring that separation into a ComfyUI workflow. It splits a soundtrack into **music, sound effects, and speech**, then lets you save the parts or combine them again. The repository includes a workflow that writes a new video using the music and effects, with the extracted dialogue saved separately.\n\nIt runs locally on a CPU or NVIDIA CUDA GPU. The two available models total about **23 MB** to download.\n\nThis is an **unofficial integration** of [Dynamic Band Splitting (DBS)](https://github.com/JusperLee/Dynamic-Band-Splitting), by Bingzhe Zhang, Kai Li, and Xiaolin Hu. The research and trained models are theirs; my contribution is the ComfyUI integration, workflows, and verification.\n\nDBS produces three audio outputs, often called stems:\n\nFor dialogue removal, the workflow adds **music + SFX** and uses that as the video's new soundtrack. For dubbing preparation or transcription, you can save the speech output on its own. You can also use the separate tracks to adjust their relative levels in a later editing step.\n\nThese are estimated sources. Some dialogue can remain in the other tracks, and some wanted sounds can be affected. Start with a short excerpt and listen to every output before processing a longer video.\n\n*A 12-second excerpt from Tears of Steel and the DBS-TIGER stems produced by this integration on CPU. Source film: Blender Foundation, [CC BY 3.0](https://creativecommons.org/licenses/by/3.0/). The excerpt comes from the [DBS project page](https://cslikai.cn/Dynamic-Band-Splitting/).*\n\nThe spectrogram shows the outputs, but listening is the useful test. The authors provide [audio examples and a listening demo](https://cslikai.cn/Dynamic-Band-Splitting/) for exploring the underlying model. Those examples are separate from the ComfyUI integration.\n\nYou need **ComfyUI v0.38.0 or later**, with the V3 node API, and **Python 3.10 or later**. A CUDA GPU is optional.\n\nRun these commands from your **ComfyUI root directory**, using the Python environment that runs ComfyUI:\n\n```\ngit clone https://github.com/hiroki-abe-58/ComfyUI-DBS custom_nodes/ComfyUI-DBS\npython -m pip install -r custom_nodes/ComfyUI-DBS/requirements.txt\n```\n\nIf you use a portable installation, replace `python` with that installation's Python executable. Installing dependencies into a different Python environment will not make them available to ComfyUI.\n\nDownload the two models:\n\n```\npython custom_nodes/ComfyUI-DBS/tools/download_model.py --models-dir models/dbs\n```\n\nThe download script uses pinned model revisions and checks the files' SHA-256 hashes. It creates these folders:\n\n```\nComfyUI/\n  models/\n    dbs/\n      DBS-TIGER-DnR/\n      DBS-BSRNN-DnR/\n```\n\nThe requirements are `einops` and `safetensors`, which ComfyUI usually already includes. After setup, the nodes load local model files; they do not download models during a run.\n\nRestart ComfyUI to load the new nodes.\n\nOpen the supplied [video_remove_dialogue.json workflow](https://github.com/hiroki-abe-58/ComfyUI-DBS/blob/main/workflows/gui/video_remove_dialogue.json) from:\n\n```\ncustom_nodes/ComfyUI-DBS/workflows/gui/video_remove_dialogue.json\n```\n\nChoose your video in **Load Video**. The workflow's `clip.mp4` filename is a placeholder for your own file.\n\nThe workflow performs these steps:\n\n`add`, combines the music and SFX.\nThe supplied workflow uses `channels = each channel` to process the stereo channels independently, and `output_sample_rate = input` to return audio at the input sample rate. The model itself works at 44.1 kHz; the node handles resampling.\n\nFor an audio-only input, the [GUI workflow folder](https://github.com/hiroki-abe-58/ComfyUI-DBS/tree/main/workflows/gui) also includes:\n\n`separate_soundtrack.json`: saves all three stems as FLAC.` remove_dialogue.json`: saves music + SFX, plus the separate dialogue track.\nYou can begin with these workflows and connect the audio outputs to your existing graph.\n\n**DBS-TIGER-DnR** is the default. It has 1.0 million parameters and higher reported separation scores than DBS-BSRNN on the authors' DnR v2 evaluation.\n\n**DBS-BSRNN-DnR** has 4.7 million parameters, but it was faster in the repository's local measurements. It is the practical starting point for CPU use.\n\nHere are two examples from the [recorded verification runs](https://github.com/hiroki-abe-58/ComfyUI-DBS/blob/main/docs/VERIFICATION.md):\n\nThese measurements used Windows 11, an already-loaded model, and the pack's runtime in-process. They exclude model loading and video export. They are useful reference points, not speed promises for another machine.\n\nStereo takes more work because each channel is processed separately. For the same 10-minute input on the RTX 5090, the recorded stereo times were **74.5 seconds for DBS-TIGER** and **53.7 seconds for DBS-BSRNN**. Use those figures when considering the stereo video workflow above.\n\nWith the loader set to `auto`, the pack uses ComfyUI's GPU when it is CUDA and otherwise falls back to CPU. Apple and AMD GPU acceleration are not implemented in this pack.\n\nA model integration can load successfully while changing the model's behavior. I wanted to check the audio output against the official package as well as make the nodes work.\n\nThe [verification report](https://github.com/hiroki-abe-58/ComfyUI-DBS/blob/main/docs/VERIFICATION.md) records:\n\nThose checks establish integration fidelity under the tested conditions. They do not establish perfect separation quality, or byte-identical output for every exported file and configuration. The report documents the export rounding and CUDA differences too.\n\nThe released models were trained on **DnR v2**, which combines English speech, music, and sound effects into film-like mixtures. Songs with sung vocals, other languages, or unusually loud music need their own evaluation. Singing may be assigned to the music track.\n\nFor a first test, choose a short clip with clearly audible dialogue, music, and effects. Listen to the original, the extracted speech, and the recombined music + SFX. That will tell you much more about your use case than a spectrogram alone.\n\n**[Try ComfyUI-DBS and the included workflows on GitHub](https://github.com/hiroki-abe-58/ComfyUI-DBS).** If it is useful in your workflow, a star helps other ComfyUI users discover it. Feedback with your model choice, device, and the kind of audio you tried is welcome in the repository's issues.\n\n*ComfyUI-DBS is independent of the DBS authors and Comfy Org. The integration code is MIT-licensed; the vendored DBS code and model licensing are documented in the repository's [license notes](https://github.com/hiroki-abe-58/ComfyUI-DBS#licenses).*", "url": "https://wpnews.pro/news/separate-dialogue-music-and-sound-effects-in-comfyui", "canonical_source": "https://dev.to/genelab_999/separate-dialogue-music-and-sound-effects-in-comfyui-apa", "published_at": "2026-10-21 01:15:47+00:00", "updated_at": "2026-10-21 01:19:44.263537+00:00", "lang": "en", "topics": ["ai-tools", "developer-tools", "artificial-intelligence"], "entities": ["ComfyUI", "ComfyUI-DBS", "Dynamic Band Splitting", "Bingzhe Zhang", "Kai Li", "Xiaolin Hu", "Blender Foundation"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/separate-dialogue-music-and-sound-effects-in-comfyui", "markdown": "https://wpnews.pro/news/separate-dialogue-music-and-sound-effects-in-comfyui.md", "text": "https://wpnews.pro/news/separate-dialogue-music-and-sound-effects-in-comfyui.txt", "jsonld": "https://wpnews.pro/news/separate-dialogue-music-and-sound-effects-in-comfyui.jsonld"}}