Method to disrupt generation quality of deep audio generation models

A system comprising:

• a memory;
• and
• a processing device operatively coupled to the memory, the processing device to perform operations comprising:
• segmenting an audio signal into a plurality of audio signal segments;
• generating a plurality of modified audio signal segments based on processing data from the plurality of audio signal segments using a trained machine learning model, wherein the plurality of modified audio signal segments are indistinguishable from the plurality of audio signal segments to an average human listener;
• and
• generating a reconstructed audio signal corresponding to the audio signal by combining the plurality of modified audio signal segments, wherein the reconstructed audio signal is indistinguishable from the audio signal to the average human listener but, when used as a seed provided to a trained generative machine learning model, constrains an ability of the trained generative machine learning model to generate new audio signals similar to the audio signal.

A non-transitory computer readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

• segmenting an audio signal into a plurality of audio signal segments;
• generating a plurality of modified audio signal segments based on processing data from the plurality of audio signal segments using a trained machine learning model, wherein the plurality of modified audio signal segments are indistinguishable from the plurality of audio signal segments to an average human listener;
• and
• generating a reconstructed audio signal corresponding to the audio signal by combining the plurality of modified audio signal segments, wherein the reconstructed audio signal is indistinguishable from the audio signal to the average human listener but, when used to train a generative machine learning model, constrains an ability of the trained generative machine learning model to generate new audio signals similar to the audio signal.