https://github.com/yoyololicon/kazane

Simple sinc interpolation in PyTorch.
https://github.com/yoyololicon/kazane

downsampling dsp pytorch-implementation upsample

Last synced: 21 days ago
JSON representation

Simple sinc interpolation in PyTorch.

• Host: GitHub
• URL: https://github.com/yoyololicon/kazane
• Owner: yoyololicon
• License: mit
• Created: 2021-06-13T11:17:23.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2023-07-08T15:59:06.000Z (over 1 year ago)
• Last Synced: 2024-10-04T15:48:40.569Z (about 1 month ago)
• Topics: downsampling, dsp, pytorch-implementation, upsample
• Language: Python
• Homepage:
• Size: 46.9 KB
• Stars: 14
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Kazane: simple sinc interpolation for 1D signal in PyTorch

[![build](https://github.com/yoyololicon/kazane/actions/workflows/python-package.yml/badge.svg)](https://github.com/yoyololicon/kazane/actions/workflows/python-package.yml)

[![Upload Python Package](https://github.com/yoyololicon/kazane/actions/workflows/python-publish.yml/badge.svg)](https://github.com/yoyololicon/kazane/actions/workflows/python-publish.yml)

[![PyPI version](https://badge.fury.io/py/kazane.svg)](https://badge.fury.io/py/kazane)

Kazane utilize FFT based convolution to provide fast sinc interpolation for 1D signal when your sample rate only needs to change by an integer amounts; If you need to change by a fraction amounts, checkout [julius](https://github.com/adefossez/julius).

## Installation

```commandline

pip install kazane

```

or 

``` commandline

pip install git+https://github.com/yoyololicon/kazane

```

for latest version.

## Usage

```python

import kazane

import torch

signal = torch.randn(8, 2, 44100)

# downsample by an amount of 3

decimater = kazane.Decimate(3)

resampled_signal = decimater(signal)

# upsample by an amount of 2

upsampler = kazane.Upsample(2)

resampled_signal = upsampler(signal)

# you can also control number of zeros, roll-off frequency of the sinc interpolation kernel

decimater = kazane.Decimate(3, num_zeros=24, roll_off=0.9)

# use other types of window function for the sinc kernel

upsampler = kazane.Upsample(2, window_func=torch.blackman_window)

```

## Benchmarks on CUDA

Using the benchmark scripts at [bench](./bench), you can see that FFT can gives some speed improvements when the sample rate changes with some common integer numbers.

```

[---------- Down sample ----------]

               |  julius  |  kazane

2 threads: ------------------------

      rate: 2  |   52.2   |   52.4 

      rate: 3  |   66.5   |   36.1 

      rate: 5  |   94.8   |   30.0 

      rate: 7  |  121.7   |   42.3 

Times are in milliseconds (ms).

[----------- Up sample -----------]

               |  julius  |  kazane

2 threads: ------------------------

      rate: 2  |   48.8   |   39.0 

      rate: 3  |   68.1   |   51.6 

      rate: 5  |  112.5   |   78.9 

      rate: 7  |  159.4   |  108.0 

Times are in milliseconds (ms).

```