DrumGAN - Adversarial synthesis of drum sounds
(2021) STAN40 20201Department of Statistics
- Abstract
- This paper faces the problem of audio synthesis with Generative Adversarial Networks (GAN), with an attempt to create novel, original high-quality samples of drums that could be used within the realm of music production.
My results show that it is possible to create high-quality drum samples with architecture such as DCGAN (Deep Convolutional GAN) and that using causal and dilated convolutions is a viable approach, while not making clear if this approach is significantly better than the one of using standard convolutions. It also shows that 1-d transpose convolutions can be substituted with nearest neighbour upsampling followed by regular 1-d convolutions for GANs that generate one-dimensional data. In the final discussion the idea of... (More) - This paper faces the problem of audio synthesis with Generative Adversarial Networks (GAN), with an attempt to create novel, original high-quality samples of drums that could be used within the realm of music production.
My results show that it is possible to create high-quality drum samples with architecture such as DCGAN (Deep Convolutional GAN) and that using causal and dilated convolutions is a viable approach, while not making clear if this approach is significantly better than the one of using standard convolutions. It also shows that 1-d transpose convolutions can be substituted with nearest neighbour upsampling followed by regular 1-d convolutions for GANs that generate one-dimensional data. In the final discussion the idea of initializing generator weights in a strategic way in order to increase GAN training stability is introduced. (Less) - Popular Abstract
- This paper faces the problem of audio synthesis with Generative Adversarial Networks (GAN), with an attempt to create novel, original high-quality samples of drums that could be used within the realm of music production.
My results show that it is possible to create high-quality drum samples with architecture such as DCGAN (Deep Convolutional GAN) and that using causal and dilated convolutions is a viable approach, while not making clear if this approach is significantly better than the one of using standard convolutions. It also shows that 1-d transpose convolutions can be substituted with nearest neighbour upsampling followed by regular 1-d convolutions for GANs that generate one-dimensional data. In the final discussion the idea of... (More) - This paper faces the problem of audio synthesis with Generative Adversarial Networks (GAN), with an attempt to create novel, original high-quality samples of drums that could be used within the realm of music production.
My results show that it is possible to create high-quality drum samples with architecture such as DCGAN (Deep Convolutional GAN) and that using causal and dilated convolutions is a viable approach, while not making clear if this approach is significantly better than the one of using standard convolutions. It also shows that 1-d transpose convolutions can be substituted with nearest neighbour upsampling followed by regular 1-d convolutions for GANs that generate one-dimensional data. In the final discussion the idea of initializing generator weights in a strategic way in order to increase GAN training stability is introduced. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9055621
- author
- Falini, Victor LU
- supervisor
- organization
- course
- STAN40 20201
- year
- 2021
- type
- H1 - Master's Degree (One Year)
- subject
- keywords
- Generative Adversarial Network, Audio Generation, Audio Synthesis, Drums, Samples, GAN, DCGAN
- language
- English
- id
- 9055621
- date added to LUP
- 2022-06-03 09:31:24
- date last changed
- 2022-06-03 11:41:28
@misc{9055621,
abstract = {{This paper faces the problem of audio synthesis with Generative Adversarial Networks (GAN), with an attempt to create novel, original high-quality samples of drums that could be used within the realm of music production.
My results show that it is possible to create high-quality drum samples with architecture such as DCGAN (Deep Convolutional GAN) and that using causal and dilated convolutions is a viable approach, while not making clear if this approach is significantly better than the one of using standard convolutions. It also shows that 1-d transpose convolutions can be substituted with nearest neighbour upsampling followed by regular 1-d convolutions for GANs that generate one-dimensional data. In the final discussion the idea of initializing generator weights in a strategic way in order to increase GAN training stability is introduced.}},
author = {{Falini, Victor}},
language = {{eng}},
note = {{Student Paper}},
title = {{DrumGAN - Adversarial synthesis of drum sounds}},
year = {{2021}},
}