Audio library for modelling loudness

Dominic Ward

Last update: Oct 2, 2022

Related tags

Audio loudness

Overview

Loudness

Loudness is a C++ library with Python bindings for modelling perceived loudness. The library consists of processing modules which can be cascaded to form a loudness model.

Dependencies

To build the C++ library you will need:

libsndfile1-dev >= 1.0.25
libfftw3-dev >= 3.3.3
zlib1g-dev >= 1.2.8

To build the Python bindings you will need:

swig >= 3.0.0
python-numpy-dev

Note

This project is still in heavy development so is not stable. I am also now only supporting Python 3.5+. Please register an issue at: https://github.com/deeuu/loudness/issues

Acknowledgments

The library interface is based on the fantastic AIM-C: https://code.google.com/p/aimc/

The cnpy library for reading numpy arrays in C++: https://github.com/rogersce/cnpy

Ricard Marxer for the loudia audio project: https://github.com/rikrd/loudia

Example - Loudness of a 1 kHz tone @ 40 dB SPL according to ANSI S3.4:2007

import loudness as ln

# All inputs and outputs make use of a SignalBank
inputBank = ln.SignalBank()
nSources = 1
nEars = 1
nChannels = 1
nSamples = 1
fs = 1

# There are 4 dimensions
inputBank.initialize(nSources, nEars, nChannels, nSamples, fs)

# Set the centre frequency of the first channel
inputBank.setCentreFreq(0, 1000)

# Set the intensity in normalised units
level = 40
inputBank.setSample(0, 0, 0, 0, 10.0 ** (level / 10.0))

# The loudness model
model = ln.StationaryLoudnessANSIS342007()
model.initialize(inputBank)

# Now process the input
model.process(inputBank)

# Get the output of this loudness model
feature = 'Loudness'
outputBank = model.getOutput(feature)

print 'Loudness in sones %0.2f' % outputBank.getSample(0, 0, 0, 0)

a library for audio and music analysis

aubio aubio is a library to label music and sounds. It listens to audio signals and attempts to detect events. For instance, when a drum is hit, at wh

2.9k Dec 30, 2022

C++ library for audio and music analysis, description and synthesis, including Python bindings

Essentia Essentia is an open-source C++ library for audio analysis and audio-based music information retrieval released under the Affero GPL license.

Music Technology Group - Universitat Pompeu Fabra

2.3k Jan 3, 2023

LibXtract is a simple, portable, lightweight library of audio feature extraction functions.

LibXtract LibXtract is a simple, portable, lightweight library of audio feature extraction functions. The purpose of the library is to provide a relat

215 Nov 16, 2022

A library for augmenting annotated audio data

muda A library for Musical Data Augmentation. muda package implements annotation-aware musical data augmentation, as described in the muda paper. The

214 Nov 22, 2022

Python audio and music signal processing library

madmom Madmom is an audio signal processing library written in Python with a strong focus on music information retrieval (MIR) tasks. The library is i

1k Dec 26, 2022

pedalboard is a Python library for adding effects to audio.

pedalboard is a Python library for adding effects to audio. It supports a number of common audio effects out of the box, and also allows the use of VST3® and Audio Unit plugin formats for third-party effects.

3.9k Jan 2, 2023

Audio fingerprinting and recognition in Python

dejavu Audio fingerprinting and recognition algorithm implemented in Python, see the explanation here: How it works Dejavu can memorize audio by liste

6k Jan 6, 2023

kapre: Keras Audio Preprocessors

Kapre Keras Audio Preprocessors - compute STFT, ISTFT, Melspectrogram, and others on GPU real-time. Tested on Python 3.6 and 3.7 Why Kapre? vs. Pre-co

867 Dec 29, 2022

?️ Open Source Audio Matching and Mastering

Matching + Mastering = ❤️ Matchering 2.0 is a novel Containerized Web Application and Python Library for audio matching and mastering. It follows a si

781 Jan 5, 2023

Comments

Can't compile on OS X

Compiling using Apple LLVM version 7.0.0 (clang-700.1.76) results in the following error:

Compiling:  ../src/support/Timer.cpp
In file included from ../src/support/Timer.cpp:20:
../src/support/Timer.h:75:9: error: unknown type name 'clockid_t'; did you mean 'clock_t'?
        clockid_t id_;
        ^~~~~~~~~
        clock_t
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.11.sdk/usr/include/sys/_types/_clock_t.h:30:33: note: 'clock_t'
      declared here
typedef __darwin_clock_t        clock_t;
                                ^
../src/support/Timer.cpp:27:19: error: use of undeclared identifier 'CLOCK_PROCESS_CPUTIME_ID'
            id_ = CLOCK_PROCESS_CPUTIME_ID;
                  ^
../src/support/Timer.cpp:29:19: error: use of undeclared identifier 'CLOCK_MONOTONIC'
            id_ = CLOCK_MONOTONIC;
                  ^
../src/support/Timer.cpp:37:9: error: use of undeclared identifier 'clock_gettime'
        clock_gettime(id_, &time1_);
        ^
../src/support/Timer.cpp:44:9: error: use of undeclared identifier 'clock_gettime'
        clock_gettime(id_, &time2_);
        ^
../src/support/Timer.cpp:67:9: error: use of undeclared identifier 'clock_getres'
        clock_getres(id_, &res);
        ^
6 errors generated.
make: *** [../src/support/Timer.o] Error 1

I am not really familiar with any of the time.h stuff but it looks like this is probably just a Mac OS X issue: http://stackoverflow.com/questions/5167269/clock-gettime-alternative-in-mac-os-x/5167506#5167506

opened by faroit 5

Get Segmentation fault (core dumped)

Thanks for the work.

After installation, I run the following sample code

import loudness as ln

# All inputs and outputs make use of a SignalBank
inputBank = ln.SignalBank()
nSources = 1
nEars = 1
nChannels = 1
nSamples = 1
fs = 1

# There are 4 dimensions
inputBank.initialize(nSources, nEars, nChannels, nSamples, fs)

# Set the centre frequency of the first channel
inputBank.setCentreFreq(0, 1000)

# Set the intensity in normalised units
level = 40
inputBank.setSample(0, 0, 0, 0, 10.0 ** (level / 10.0))

# The loudness model
model = ln.StationaryLoudnessANSIS342007()
model.initialize(inputBank)

# Now process the input
model.process(inputBank)

# Get the output of this loudness model
feature = 'InstantaneousLoudness'
outputBank = model.getOutput(feature)

print('Loudness in sones %0.2f' % outputBank.getSample(0, 0, 0, 0))

And got the following messages

/home/sen/env/ml/lib/python3.6/site-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
Segmentation fault (core dumped)

It seems the problem occurs at

outputBank.getSample(0, 0, 0, 0)

Do you have an idea of this problem?

I have tried python 2.7 and python 3.6. OS: ubuntu 16.04.6 gcc: 5.4.0 g++: 5.4.0

Thanks

opened by lightondust 4

Can't build python bindings on ubuntu

Run the following setup commands:

cd build
make
make install
cd ../swig
python setup.py install

I get the following:

sudo python setup.py install
/usr/lib/python2.7/distutils/dist.py:267: UserWarning: Unknown distribution option: 'install_requires'
  warnings.warn(msg)
running install
running build
running build_py
running build_ext
building 'loudness._core' extension
swigging core.i to core_wrap.cpp
swig -python -c++ -o core_wrap.cpp core.i
../src/thirdParty/cnpy/cnpy.h:236: Warning 302: Identifier 'create_npy_header' redefined (ignored),
../src/thirdParty/cnpy/cnpy.h:40: Warning 302: previous definition of 'create_npy_header'.
../src/support/Model.h:170: Error: Syntax error in input(3).
error: command 'swig' failed with exit status 1

opened by nickjillings 4

Can't compile on Ubuntu 14.04

Hi, thanks a lot for the work ! But I can't manage to install it properly... Do we agree that I only have to go to the swig folder and do ./build.sh in a shell ?

I try to do so on a Ubuntu 14.04 system with swig3.0 installed with a symbolic link as swig (apt-get install swig wants to install swig2.0) but it's not ending well, core_wrap.cpp seems to not be created.

Do you have an idea of the problem ? Does it need python 3 ? I'm using python 2.7 thank you, Raph

opened by raphleiba 2