Audio Analysis

👁️ Yvann Nodes/🔊 Audio

Analyzes audio to generate reactive weights and visualization graphs. Can extract specific elements like drums, vocals, or bass from the audio. Input requires an audio_sep_model from Load Audio Separation Model, the audio file, batch_size (number of frames), and fps (frames per second). Choose an analysis_mode to select which audio component to analyze. The threshold parameter sets the minimum weight value to pass through (0.0 to 1.0), and multiply amplifies the weights before normalization. Outputs include audio_weights (list of reactive weights), processed_audio (separated component), original_audio (unchanged input), and graph_audio (visual representation).

Pack: ComfyUI_Yvann-Nodes

custom_nodes.ComfyUI_Yvann-Nodes

Inputs (7)

Name	Type	Required
audio_sep_model	AUDIO_SEPARATION_MODEL	required
batch_size	INT	required
fps	FLOAT	required
audio	AUDIO	required
analysis_mode	COMBO	required
threshold	FLOAT	required
multiply	FLOAT	required

Outputs (4)

Name	Type
processed_audio	AUDIO
original_audio	AUDIO
audio_weights	FLOATS
graph_audio	IMAGE