• Computer Science
  • Published in ICLR 2016

SampleRNN: An Unconditional End-to-End Neural Audio Generation Model

@article{Mehri2016SampleRNNAU,
  title={SampleRNN: An Unconditional End-to-End Neural Audio Generation Model},
  author={Soroush Mehri and Kundan Kumar and Ishaan Gulrajani and Rithesh Kumar and Shubham Jain and Jose Sotelo and Aaron C. Courville and Yoshua Bengio},
  journal={ArXiv},
  year={2016},
  volume={abs/1612.07837}
}
In this paper we propose a novel model for unconditional audio generation task that generates one audio sample at a time. We show that our model which profits from combining memory-less modules, namely autoregressive multilayer perceptron, and stateful recurrent neural networks in a hierarchical structure is de facto powerful to capture the underlying sources of variations in temporal domain for very long time on three datasets of different nature. Human evaluation on the generated samples… CONTINUE READING

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 213 CITATIONS

B-Spline Pdf: A Generalization of Histograms to Continuous Density Models for Generative Audio Networks

  • Yannis Agiomyrgiannakis
  • Computer Science
  • 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2018
VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Digital Audio Generation with Neural Networks

VIEW 9 EXCERPTS
CITES METHODS, BACKGROUND & RESULTS
HIGHLY INFLUENCED

Generating Albums with SampleRNN to Imitate Metal, Rock, and Punk Bands

  • Dadabots, emperorcj
  • 2018
VIEW 6 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Deep Audio-Visual Learning: A Survey

VIEW 4 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

A Real-Time Wideband Neural Vocoder at 1.6 kb/s Using LPCNet

VIEW 6 EXCERPTS
CITES METHODS
HIGHLY INFLUENCED

Acoustic Scene Generation with Conditional Samplernn

VIEW 8 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Dnn-based Spectral Enhancement for Neural Waveform Generators with Low-bit Quantization

VIEW 6 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Learning Bandwidth Expansion Using Perceptually-motivated Loss

VIEW 4 EXCERPTS
CITES METHODS
HIGHLY INFLUENCED

Dynamic Spectrum Assignment for Land Mobile Radio with Deep Recurrent Neural Networks

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Generating Black Metal and Math Rock: Beyond Bach, Beethoven, and Beatles

VIEW 5 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2017
2020

CITATION STATISTICS

  • 52 Highly Influenced Citations

  • Averaged 69 Citations per year from 2017 through 2019

References

Publications referenced by this paper.
SHOWING 1-10 OF 28 REFERENCES

Pixel Recurrent Neural Networks

VIEW 19 EXCERPTS
HIGHLY INFLUENTIAL

WaveNet: A Generative Model for Raw Audio

VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

A Clockwork RNN

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

Adam: A Method for Stochastic Optimization

VIEW 1 EXCERPT
HIGHLY INFLUENTIAL

Long short-term memory in recurrent neural networks

VIEW 2 EXCERPTS
HIGHLY INFLUENTIAL

Learning Complex, Extended Sequences Using the Principle of History Compression

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL