comfyui-audio-plugin

A collection of nodes for working with audio data.

• Audio Waveforms
• Spectrograms
  • Complex-Valued Spectrograms
  • Real-Valued Spectrograms
• Filter Banks
  • Linear-scale Spectrograms
  • Mel-scale Spectrograms
• License

Note

This plugin is compatible with the ComfyUI audio nodes.

Audio Waveforms

Name	Description
Load Audio From Path	Same as "Load Audio" but loads from a local path instead of an uploaded audio file.
Plot Waveform	Plots the waveform of an AUDIO object.

Spectrograms

Name	Description
Spectrogram	Computes the spectrogram of a given AUDIO object.
Inverse Spectrogram	Converts a complex-valued spectrogram to AUDIO.
Griffin Lim	Converts a real-valued spectrogram to AUDIO using the Griffin-Lim algorithm.
Plot Spectrogram	Plots the spectrogram of a SPECT object.

A spectrogram (SPECT) can be one of three types:

1. complex – a complex number valued spectrogram;
2. magnitude – a real valued spectrogram that is the absolute value of the complex spectrogram (power 1);
3. power – a real valued spectrogram that is the square of the magnitude spectrogram (power 2);

Complex-Valued Spectrograms

Real-Valued Spectrograms

Filter Banks

Name	Description
Linear Filter Bank	Create a linear filter bank.
Mel-scale Filter Bank	Create a mel-scale filter bank.
Apply Filter Bank	Apply the specified filter bank to the spectrogram.
Plot Filter Bank	Plots the FILTER_BANK object.

Linear-scale Spectrograms

Mel-scale Spectrograms

License

Copyright (C) 2024 Reece H. Dunn

SPDX-License-Identifier: GPL-3