Commit 8d0be90b authored by Anthony Larcher's avatar Anthony Larcher
Browse files

add-on beat

parent 2f28db99
Diarization lib
+++++++++++++++
=============================
S4D - Sidekit For Diarization
=============================
| **S4D** is an open source package for speaker diarization.
| The aim of **S4D** is to provide an educational and efficient toolkit
| for speaker diarization including the whole chain of treatment
| that goes from the audio data to the analysis of the system performance.
PREREQUISITES
=============
*Sidekit for Diarization* requires the following software installed for your platform:
1. [Python](http://www.python.org)
2. [NumPy](http://www.numpy.org/)
3. [Scipy](http://http://www.scipy.org/)
4. [Pandas](http://http://www.pandas.org/)
5. [GLPK](https://www.gnu.org/software/glpk/)
6. [Sphinx 1.1.0 or newer](http://http://sphinx-doc.org/) to build the documentation
7. [SIDEKIT](https://lium.univ-lemans.fr/sidekit/)
INSTALLATION
============
We recommend the use of a virtual environment (e.g. [Miniconda](https://conda.io/miniconda.html) or [Virtualenv](https://virtualenv.readthedocs.io/en/latest/)).
TUTORIALS
=========
Once your installation is complete, you can take a look at the [tutorials](https://git-lium.univ-lemans.fr/Meignier/s4d/tree/master/tutorials).
:Authors:
Sylvain Meignier \&
Anthony Larcher \&
Pierre-Alexandre Broux \&
Florent Desnous
:Version: 0.1.0 of 2020/01/23
\ No newline at end of file
......@@ -57,4 +57,4 @@ __maintainer__ = "Sylvain Meignier"
__email__ = "sylvain.meignierr@univ-lemans.fr"
__status__ = "Production"
__docformat__ = 'reStructuredText'
__version__ = "0.1.0"
__version__ = "0.1.4.1"
......@@ -105,6 +105,25 @@ class ModelIV:
return self.ivectors
def train_per_segment(self, feature_server, idmap, normalization=True):
"""
:param feature_server:
:param idmap:
:param normalization:
:return:
"""
stat = StatServer(idmap, distrib_nb=self.ubm.distrib_nb(), feature_size=self.ubm.dim())
stat.accumulate_stat(ubm=self.ubm, feature_server=feature_server, seg_indices=range(stat.segset.shape[0]),
num_thread=self.nb_thread)
fa = FactorAnalyser(mean=self.tv_mean, Sigma=self.tv_sigma, F=self.tv)
self.ivectors = fa.extract_ivectors_single(self.ubm, stat)
if normalization:
self.ivectors.spectral_norm_stat1(self.norm_mean[:1], self.norm_cov[:1])
return self.ivectors
def score_cosine(self, use_wccn=True):
"""
......
......@@ -5,8 +5,8 @@ scipy>=0.19.0
matplotlib>=2.0.2
bottleneck>=1.3.1
setuptools>=38.5.2
sidekit>=1.3
sidekit>=1.3.>6.4
six>=1.11.0
scikit_learn>=0.19.1
sortedcontainers>=1.5.9
h5py>=2.5.0
\ No newline at end of file
h5py>=2.5.0
......@@ -33,7 +33,7 @@ setup(
license='LGPL',
platforms=['Linux, Windows', 'MacOS'],
description='S4D: SIDEKIT for Diarization',
long_description=open('README.rst').read(),
long_description=open('README.txt').read(),
install_requires=[
"mock>=1.0.1",
"nose>=1.3.4",
......@@ -48,12 +48,12 @@ setup(
"PyYAML>=3.11",
"h5py>=2.5.0",
"pandas>=0.21.1",
"PyAudio==0.2.11",
"PyAudio>=0.2.11",
"bottleneck>=1.3.1",
"setuptools>=38.5.2",
"sidekit>1.3",
"scikit_learn==0.19.1",
"sortedcontainers==1.5.9"
"sidekit>=1.3.6.4",
"scikit_learn>=0.22",
"sortedcontainers>=1.5.9"
],
package_data={'s4d': ['docs/*']},
classifiers=[ "Programming Language :: Python :: 3",
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment