loongson/pypi/: soxr-0.3.7 metadata and description

High quality, one-dimensional sample-rate conversion library

author	Myungchul Keum
classifiers	Development Status :: 4 - Beta Intended Audience :: Developers Intended Audience :: Science/Research Intended Audience :: Telecommunications Industry Topic :: Multimedia :: Sound/Audio :: Analysis Topic :: Multimedia :: Sound/Audio :: Conversion Topic :: Scientific/Engineering License :: OSI Approved :: GNU Lesser General Public License v2 or later (LGPLv2+) Programming Language :: C Programming Language :: Cython Programming Language :: Python :: 3
description_content_type	text/markdown
keywords	audio resampling,samplerate conversion,SRC,signal processing,resampler
license	LGPLv2.1+
project_urls	Documentation, https://python-soxr.readthedocs.io Source, https://github.com/dofuuz/python-soxr Bug Tracker, https://github.com/dofuuz/python-soxr/issues
provides_extras	test
requires_dist	numpy sphinx; extra == "docs" sphinx-book-theme; extra == "docs" myst-parser; extra == "docs" linkify-it-py; extra == "docs" pytest; extra == "test"
requires_python	>=3.6

Because this project isn't in the mirror_whitelist, no releases from root/pypi are included.

File	Tox results	History
soxr-0.3.7-cp38-cp38-linux_loongarch64.whl Size 407 KB Type Python Wheel Python 3.8		Uploaded to loongson/pypi by loongson 2024-09-09 09:05:47

Python-SoXR

High quality, one-dimensional sample-rate conversion library for Python.

Homepage: https://github.com/dofuuz/python-soxr
Documentation: https://python-soxr.readthedocs.io
PyPI: https://pypi.org/project/soxr/

Keywords: Resampler, Audio resampling, Samplerate conversion, DSP(Digital Signal Processing)

Python-SoXR is a Python wrapper of libsoxr.

Installation

pip install soxr

If installation fails, upgrade pip with python -m pip install --upgrade pip and try again.

in Conda environment

conda install -c conda-forge soxr-python

Note: Conda packge name is soxr-python, not python-soxr.

Basic usage

import soxr

y = soxr.resample(
    x,          # 1D(mono) or 2D(frames, channels) array input
    48000,      # input samplerate
    16000       # target samplerate
)

If input is not numpy.ndarray, it will be converted to numpy.ndarray(dtype='float32').
dtype should be one of float32, float64, int16, int32.

Output is numpy.ndarray with same dimension and data type of input.

Streaming usage

Use ResampleStream for real-time processing or very long signal.

import soxr

rs = soxr.ResampleStream(
    44100,              # input samplerate
    16000,              # target samplerate
    1,                  # channel(s)
    dtype='float32'     # data type (default = 'float32')
)

eof = False
while not eof:
    # Get chunk
    ...

    y_chunk = rs.resample_chunk(
        x,              # 1D(mono) or 2D(frames, channels) array input
        last=eof        # Set True at end of input
    )

Output frame count may not be consistent. This is normal operation.
(ex. [0, 0, 0, 186, 186, 166, 186, 186, 168, ...])

Requirement

x86 and ARM processors are supported.

Neon extension is required for ARM CPUs. Without Neon, it may crash or return inaccurate result.

Benchmark

Sweep, impulse, speed compairsion with other Python resamplers.

https://colab.research.google.com/drive/1XgSOvWlRIau1FYwQG_yRSAhDK3KB8bEL?usp=sharing

Speed comparison summary

Downsampling 10 sec of 48000 Hz to 44100 Hz.
Ran on Google Colab.

Library	Time on CPU (ms)
soxr (HQ)	7.2
scipy.signal.resample	13.4
soxr (VHQ)	15.8
torchaudio	19.2
lilfilter	21.4
julius	23.1
resampy (kaiser_fast)	62.6
samplerate (sinc_medium)	92.5
resampy (kaiser_best)	256
samplerate (sinc_best)	397

Technical detail

For technical details behind resampler, see libsoxr docs.

Python-SoXR uses forked version of libsoxr. See difference here.
These changes does not apply to dynamic linked builds(e.g. conda-forge build)

Credit and License

Python-SoXR is LGPL v2.1+ licensed, following libsoxr's license.

OSS libraries used

libsoxr (LGPLv2.1+)

The SoX Resampler library
https://sourceforge.net/projects/soxr/