Skip to main content
This node converts a compressed audio representation (latent samples) back into an audio waveform using a Variational Autoencoder (VAE). It processes the data in smaller, overlapping sections (tiles) to manage memory usage, making it suitable for handling longer audio sequences.

Inputs

ParameterDescriptionData TypeRequiredRange
samplesThe compressed latent representation of the audio to be decoded.LATENTYesN/A
vaeThe Variational Autoencoder model used to perform the decoding.VAEYesN/A
tile_sizeThe size of each processing tile. The audio is decoded in sections of this length to conserve memory (default: 512).INTYes32 to 8192
overlapThe number of samples that adjacent tiles overlap. This helps to reduce artifacts at the boundaries between tiles (default: 64).INTYes0 to 1024

Outputs

Output NameDescriptionData Type
outputThe decoded audio waveform.AUDIO
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): d846ec48856b8a1415101b8db06b5dd38b5d1222c117ea4f170daf3347ed1685