Pruning of an Audio Enhancing Deep Generative Neural Network

Plain S, Biswas A. Pruning of an Audio Enhancing Deep Generative Neural Network. In: AES Convention 148. Audio Engineering Society; 2020. Paper 10352. Available from: https://aes.org/publications/elibrary-page/?id=20769

@inproceedings{Plain2020_20769,
  author = {Plain, Simon and Biswas, Arijit},
  title = {{Pruning of an Audio Enhancing Deep Generative Neural Network}},
  booktitle = {AES Convention 148},
  note = {Paper 10352},
  year = {2020},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=20769}
}

TY  - CPAPER
TI  - Pruning of an Audio Enhancing Deep Generative Neural Network
AU  - Plain, Simon
AU  - Biswas, Arijit
T2  - AES Convention 148
M1  - Paper 10352
PY  - 2020
DA  - 2020/05/06
UR  - https://aes.org/publications/elibrary-page/?id=20769
PB  - Audio Engineering Society
LA  - en
AB  - The performance of deep neural networks has been shown to be very effective in multiple arenas. As their use becomes more widespread, increased focus is being placed on how implementable they are on a wider variety of devices. These devices often include those with low enough processing power that the operation of the neural network has a significant impact on storage and computation resources. Here we investigate the impact of pruning weights from a deep generative convolutional auto-encoder with skip connections. The chosen model is trained in a Generative Adversarial Network (GAN) setting for audio post-processing of a low bitrate audio coder. We evaluate performance using a combination of objective scores and listening tests.
ER  -