Deep Learning & 3D Convolutional Neural Networks for Speaker Verification

Related tags

Deep Learningdeep-learningconvolutional-neural-networksspeaker-recognition3d
Overview

TensorFlow implementation of 3D Convolutional Neural Networks for Speaker Verification - Official Project Page - Pytorch Implementation

https://img.shields.io/badge/contributions-welcome-brightgreen.svg?style=flat https://badges.frapsoft.com/os/v2/open-source.svg?v=102 https://img.shields.io/twitter/follow/amirsinatorfi.svg?label=Follow&style=social

This repository contains the code release for our paper titled as "Text-Independent Speaker Verification Using 3D Convolutional Neural Networks". The link to the paper is provided as well.

The code has been developed using TensorFlow. The input pipeline must be prepared by the users. This code is aimed to provide the implementation for Speaker Verification (SR) by using 3D convolutional neural networks following the SR protocol.

readme_images/conv_gif.gif

Citation

If you used this code, please kindly consider citing the following paper:

@article{torfi2017text,
  title={Text-independent speaker verification using 3d convolutional neural networks},
  author={Torfi, Amirsina and Nasrabadi, Nasser M and Dawson, Jeremy},
  journal={arXiv preprint arXiv:1705.09422},
  year={2017}
}

DEMO

For running a demo, after forking the repository, run the following scrit:

./run.sh

speakerrecognition

General View

We leveraged 3D convolutional architecture for creating the speaker model in order to simultaneously capturing the speech-related and temporal information from the speakers' utterances.

Speaker Verification Protocol(SVP)

In this work, a 3D Convolutional Neural Network (3D-CNN) architecture has been utilized for text-independent speaker verification in three phases.

1. At the development phase, a CNN is trained to classify speakers at the utterance-level.

2. In the enrollment stage, the trained network is utilized to directly create a speaker model for each speaker based on the extracted features.

3. Finally, in the evaluation phase, the extracted features from the test utterance will be compared to the stored speaker model to verify the claimed identity.

The aforementioned three phases are usually considered as the SV protocol. One of the main challenges is the creation of the speaker models. Previously-reported approaches create speaker models based on averaging the extracted features from utterances of the speaker, which is known as the d-vector system.

How to leverage 3D Convolutional Neural Networks?

In our paper, we propose the implementation of 3D-CNNs for direct speaker model creation in which, for both development and enrollment phases, an identical number of speaker utterances is fed to the network for representing the spoken utterances and creation of the speaker model. This leads to simultaneously capturing the speaker-related information and building a more robust system to cope with within-speaker variation. We demonstrate that the proposed method significantly outperforms the d-vector verification system.

Code Implementation

The input pipeline must be provided by the user. Please refer to ``code/0-input/input_feature.py`` for having an idea about how the input pipeline works.

Input Pipeline for this work

readme_images/Speech_GIF.gif

The MFCC features can be used as the data representation of the spoken utterances at the frame level. However, a drawback is their non-local characteristics due to the last DCT 1 operation for generating MFCCs. This operation disturbs the locality property and is in contrast with the local characteristics of the convolutional operations. The employed approach in this work is to use the log-energies, which we call MFECs. The extraction of MFECs is similar to MFCCs by discarding the DCT operation. The temporal features are overlapping 20ms windows with the stride of 10ms, which are used for the generation of spectrum features. From a 0.8- second sound sample, 80 temporal feature sets (each forms a 40 MFEC features) can be obtained which form the input speech feature map. Each input feature map has the dimen- sionality of ζ × 80 × 40 which is formed from 80 input frames and their corresponding spectral features, where ζ is the number of utterances used in modeling the speaker during the development and enrollment stages.

The speech features have been extracted using [SpeechPy] package.

Implementation of 3D Convolutional Operation

The Slim high-level API made our life very easy. The following script has been used for our implementation:

net = slim.conv2d(inputs, 16, [3, 1, 5], stride=[1, 1, 1], scope='conv11')
net = PReLU(net, 'conv11_activation')
net = slim.conv2d(net, 16, [3, 9, 1], stride=[1, 2, 1], scope='conv12')
net = PReLU(net, 'conv12_activation')
net = tf.nn.max_pool3d(net, strides=[1, 1, 1, 2, 1], ksize=[1, 1, 1, 2, 1], padding='VALID', name='pool1')

############ Conv-2 ###############
############ Conv-1 ###############
net = slim.conv2d(net, 32, [3, 1, 4], stride=[1, 1, 1], scope='conv21')
net = PReLU(net, 'conv21_activation')
net = slim.conv2d(net, 32, [3, 8, 1], stride=[1, 2, 1], scope='conv22')
net = PReLU(net, 'conv22_activation')
net = tf.nn.max_pool3d(net, strides=[1, 1, 1, 2, 1], ksize=[1, 1, 1, 2, 1], padding='VALID', name='pool2')

############ Conv-3 ###############
############ Conv-1 ###############
net = slim.conv2d(net, 64, [3, 1, 3], stride=[1, 1, 1], scope='conv31')
net = PReLU(net, 'conv31_activation')
net = slim.conv2d(net, 64, [3, 7, 1], stride=[1, 1, 1], scope='conv32')
net = PReLU(net, 'conv32_activation')
# net = slim.max_pool2d(net, [1, 1], stride=[4, 1], scope='pool1')

############ Conv-4 ###############
net = slim.conv2d(net, 128, [3, 1, 3], stride=[1, 1, 1], scope='conv41')
net = PReLU(net, 'conv41_activation')
net = slim.conv2d(net, 128, [3, 7, 1], stride=[1, 1, 1], scope='conv42')
net = PReLU(net, 'conv42_activation')
# net = slim.max_pool2d(net, [1, 1], stride=[4, 1], scope='pool1')

############ Conv-5 ###############
net = slim.conv2d(net, 128, [4, 3, 3], stride=[1, 1, 1], normalizer_fn=None, scope='conv51')
net = PReLU(net, 'conv51_activation')

# net = slim.conv2d(net, 256, [1, 1], stride=[1, 1], scope='conv52')
# net = PReLU(net, 'conv52_activation')

# Last layer which is the logits for classes
logits = tf.contrib.layers.conv2d(net, num_classes, [1, 1, 1], activation_fn=None, scope='fc')

As it can be seen, slim.conv2d has been used. However, simply by using 3D kernels as [k_x, k_y, k_z] and stride=[a, b, c] it can be turned into a 3D-conv operation. The base of the slim.conv2d is tf.contrib.layers.conv2d. Please refer to official Documentation for further details.

Disclaimer

The code architecture part has been heavily inspired by Slim and Slim image classification library. Please refer to this link for further details.

Citation

If you used this code please kindly cite the following paper:

@article{torfi2017text,
  title={Text-Independent Speaker Verification Using 3D Convolutional Neural Networks},
  author={Torfi, Amirsina and Nasrabadi, Nasser M and Dawson, Jeremy},
  journal={arXiv preprint arXiv:1705.09422},
  year={2017}
}

License

The license is as follows:

APPENDIX: How to apply the Apache License to your work.

   To apply the Apache License to your work, attach the following
   boilerplate notice, with the fields enclosed by brackets "{}"
   replaced with your own identifying information. (Don't include the brackets!)  The text should be enclosed in the appropriate
   comment syntax for the file format. We also recommend that a
   file or class name and description of purpose be included on the
   same "printed page" as the copyright notice for easier
   identification within third-party archives.

Copyright {2017} {Amirsina Torfi}

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

Please refer to LICENSE file for further detail.

Contribution

We are looking forward to your kind feedback. Please help us to improve the code and make our work better. For contribution, please create the pull request and we will investigate it promptly. Once again, we appreciate your feedback and code inspections.

references

[SpeechPy] Amirsina Torfi. 2017. astorfi/speech_feature_extraction: SpeechPy. Zenodo. doi:10.5281/zenodo.810392.
Comments
  • ValueError: Convolution expects input with rank 4, got 5

    ValueError: Convolution expects input with rank 4, got 5

    When I run the run.sh, it shows something wrong:

    /home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.
  from numpy.core.umath_tests import inner1d/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.
  DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/learning_curve.py:22: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20
  DeprecationWarning)
Train data shape: (12, 80, 40, 20)Train label shape: (12,)Test data shape: (12, 80, 40, 20)
Test label shape: (12,)
Traceback (most recent call last):  File "./code/1-development/train_softmax.py", line 602, in <module>    tf.app.run()
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 125, in run
    _sys.exit(main(argv))
  File "./code/1-development/train_softmax.py", line 414, in main
    logits, end_points_speech = model_speech_fn(batch_speech[i * step: (i + 1) * step])  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/code/1-development/nets/nets_factory.py", line 59, in network_fn
    return func(images, num_classes, is_training=is_training)
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/code/1-development/nets/cnn_speech.py", line 118, in speech_cnn
    net = slim.conv2d(inputs, 16, [3, 1, 5], stride=[1, 1, 1], scope='conv11')
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/contrib/framework/python/ops/arg_scope.py", line 183, in func_with_args
    return func(*args, **current_args)
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/contrib/layers/python/layers/layers.py", line 1154, in convolution2d
    conv_dims=2)
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/contrib/framework/python/ops/arg_scope.py", line 183, in func_with_args
    return func(*args, **current_args)
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/contrib/layers/python/layers/layers.py", line 1025, in convolution
    (conv_dims + 2, input_rank))
ValueError: Convolution expects input with rank 4, got 5
Closing remaining open files:data/development_sample_dataset_speaker.hdf5...done
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
Enrollment data shape: (108, 80, 40, 1)
Enrollment label shape: (108,)
Evaluation data shape: (12, 80, 40, 1)
Evaluation label shape: (12,)
Traceback (most recent call last):
  File "./code/2-enrollment/enrollment.py", line 330, in <module>
    tf.app.run()
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 125, in run
    _sys.exit(main(argv))
  File "./code/2-enrollment/enrollment.py", line 201, in main
    for i in xrange(FLAGS.num_clones):
NameError: name 'xrange' is not defined
Closing remaining open files:data/development_sample_dataset_speaker.hdf5...donedata/enrollment-evaluation_sample_dataset.hdf5...done
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
Enrollment data shape: (108, 80, 40, 1)
Enrollment label shape: (108,)
Evaluation data shape: (12, 80, 40, 1)
Evaluation label shape: (12,)
Traceback (most recent call last):
  File "./code/3-evaluation/evaluation.py", line 380, in <module>
    tf.app.run()
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 125, in run
    _sys.exit(main(argv))
  File "./code/3-evaluation/evaluation.py", line 202, in main
    for i in xrange(FLAGS.num_clones):
NameError: name 'xrange' is not defined
Closing remaining open files:data/enrollment-evaluation_sample_dataset.hdf5...donedata/development_sample_dataset_speaker.hdf5...done
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.
  from numpy.core.umath_tests import inner1d
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.
  DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/learning_curve.py:22: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20
  DeprecationWarning)
Traceback (most recent call last):
  File "./code/4-ROC_PR_curve/calculate_roc.py", line 23, in <module>
    score = np.load(os.path.join(FLAGS.evaluation_dir,'score_vector.npy'))
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/numpy/lib/npyio.py", line 384, in load
    fid = open(file, "rb")
FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy'
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.
  from numpy.core.umath_tests import inner1d
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.
  DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/learning_curve.py:22: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20
  DeprecationWarning)
Traceback (most recent call last):
  File "./code/4-ROC_PR_curve/PlotROC.py", line 73, in <module>
    score = np.load(os.path.join(FLAGS.evaluation_dir,'score_vector.npy'))
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/numpy/lib/npyio.py", line 384, in load
    fid = open(file, "rb")
FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy'
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.
  from numpy.core.umath_tests import inner1d
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.
  DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/learning_curve.py:22: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20
  DeprecationWarning)
Traceback (most recent call last):
  File "./code/4-ROC_PR_curve/PlotPR.py", line 58, in <module>
    score = np.load(os.path.join(FLAGS.evaluation_dir,'score_vector.npy'))
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/numpy/lib/npyio.py", line 384, in load
    fid = open(file, "rb")
FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy'
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning: numpy.dtype size changed, may indicate binary incompatibility. Expected 96, got 88
  return f(*args, **kwds)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.
  from numpy.core.umath_tests import inner1d
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.
  DeprecationWarning)
/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/sklearn/learning_curve.py:22: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the functions are moved. This module will be removed in 0.20
  DeprecationWarning)
Traceback (most recent call last):
  File "./code/4-ROC_PR_curve/PlotHIST.py", line 53, in <module>
    score = np.load(os.path.join(FLAGS.evaluation_dir,'score_vector.npy'))
  File "/home/jovyan/Documents/git/3D-convolutional-speaker-recognition/speacker-rec-py35/lib/python3.5/site-packages/numpy/lib/npyio.py", line 384, in load
    fid = open(file, "rb")
FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy'
    opened by 8rV1n 42
  • Problem with evaluation.

    Problem with evaluation.

    Hi @astorfi ,thank for your great work, i also use all the same settings but use hdf5 to store training data instead of Audio Dataset. However, my evaluation result is low, EER is up to 40%. I think there is something wrong with my work. Do you have any idea to fix this? I use VoxCeleb dataset for background model and only use 1 sample per speaker. 50 people for enrollment, 50 for un-enrollment (reject). 4 samples for evaluation.

    Thank for your help.

    opened by duynguyen5896 16
  • Regarding input data

    Regarding input data

    Hi @astorfi I have gone through your code. While extracting mfcc features for sample audio file it contains shape (420,40) here, 420 is number of frames and 40 is number of features.But In sample data of your code youre applying mfec feature file contains shape (3209,40,3). As per my understanding 3209 is Number of Frames,40 is Number of Features,3 Is number of Channels. I didn't understand the number of channels usage.can you please suggest how to create Feature_mfec.npy file in your format.

    opened by abhishekkritarth 14
  • How to generate data

    How to generate data

    Hello,

    I have a dataset of voices. I want to generate development and enrollment hdf5 file. The input_feature.py file seams to generate development files (nx80x40x20). How can I generate the enrollment file?

    opened by xav12358 13
  • Retrain on new and own dataset

    Retrain on new and own dataset

    Hi,

    First, that is a great job and ver well done :) Now I am trying to use your source code and maybe contribute to it, I am working on a speaker recognition problem to detect if a teacher tutorial is recorded by his voic. I have about 10 hours of historical recordings for 6 teachers. First I used the speechpy to get 3d npy from wav files and used your create_development.py to create the hdf5 files for train and eval. Is that correct? Specially I got 13 instead of 40 regarding the features vector length in the npy files! I ran the run.bash file and it gave me also error saying something like that: ValueError: Negative dimension size caused by subtracting 2 from 1 for 'MaxPool_7' (op: 'MaxPool') with input shapes: [?,1,112,128].

    opened by Ahmed-Abouzeid 10
  • Convolution expects input with rank 4, got 5

    Convolution expects input with rank 4, got 5

    Traceback (most recent call last): File "./code/1-development/train_softmax.py", line 602, in tf.app.run() File "/opt/tensorflow/python2.7/local/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 125, in run _sys.exit(main(argv)) File "./code/1-development/train_softmax.py", line 414, in main logits, end_points_speech = model_speech_fn(batch_speech[i * step: (i + 1) * step]) File "/opt/speaker-recognition/code/1-development/nets/nets_factory.py", line 59, in network_fn return func(images, num_classes, is_training=is_training) File "/opt/speaker-recognition/code/1-development/nets/cnn_speech.py", line 118, in speech_cnn net = slim.conv2d(inputs, 16, [3, 1, 5], stride=[1, 1, 1], scope='conv11') File "/opt/tensorflow/python2.7/local/lib/python2.7/site-packages/tensorflow/contrib/framework/python/ops/arg_scope.py", line 183, in func_with_args return func(*args, **current_args) File "/opt/tensorflow/python2.7/local/lib/python2.7/site-packages/tensorflow/contrib/layers/python/layers/layers.py", line 1154, in convolution2d conv_dims=2) File "/opt/tensorflow/python2.7/local/lib/python2.7/site-packages/tensorflow/contrib/framework/python/ops/arg_scope.py", line 183, in func_with_args return func(*args, **current_args) File "/opt/tensorflow/python2.7/local/lib/python2.7/site-packages/tensorflow/contrib/layers/python/layers/layers.py", line 1025, in convolution (conv_dims + 2, input_rank)) ValueError: Convolution expects input with rank 4, got 5

    opened by hungryquiter 9
  • loss = 0,train acc = 0

    loss = 0,train acc = 0

    Hi astorfi, I'm trying to use your code to train a model with 31 labels, 60 samples for each label. However, when i use train_softmax.py, last minibatches return loss = 0 while train acc = 0. Do you have any idea to fix it?

    image

    Thank you.

    opened by duynguyen5896 8
  • shape mismatch problem in input_feature.py

    shape mismatch problem in input_feature.py

    Hello! We are trying to make our own input pipeline. However, when we follow the getitem method in Audioset (with the setting that cube_shape is (20,80,40)), there is a shape mismatch when the model tries to feed data for batch_speech (placeholder with the shape of (20,80,.40,1)).

    After carefully review the code in train_softmax.py, we find that the input shape will conflict with the transpose operation in following code:

    speech_train = np.transpose(speech_train[None, :, :, :, :], axes=(1, 4, 2, 3, 0))

    What is the solution? Could you give us any help?

    opened by yangalan123 8
  • prediction difference between batch=1 and batch=16

    prediction difference between batch=1 and batch=16

    Any ideas why I'm receiving different prediction values when running with batch_size=1,16? find code below: Thanks!

    def predict(self,speech_input): labels = np.empty(0, int) labels = np.append(labels, range(speech_input.shape[0]), axis=0) feature,logits,_ = self.session.run( [self.features,self.logits,self.end_points_speech], feed_dict={self.is_training: False, self.batch_dynamic: labels.shape[0], self.margin_imp_tensor: 50, self.batch_speech: speech_input}) #self.batch_labels: labels.reshape([labels.shape[0], 1])})

        # Extracting the associated numpy array.
    #print (feature[0])

    return  feature,logits
    opened by alanbekker 8
  • .wav inputs specifics

    .wav inputs specifics

    Hi guys, I have a question regarding the input wav files used for training. What are the audio format specifications? I used voxceleb ( http://www.robots.ox.ac.uk/~vgg/data/voxceleb/ ) as dataset, but it is giving me some troubles. Do you know about any other usable dataset?

    Thank you ;)

    opened by loregagliard 6
  • Run time error in the demo

    Run time error in the demo

    When I ran the run.sh, the execution terminated saying: FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy'

    Where do i get this score file from? Do I need to create one? I just ran the run.sh for demo.

    Can you please help?

    Regards!

    opened by sivagururaman 6
  • 　All Dependency of This Project

    　All Dependency of This Project

    Hi, astorfi First of all, thanks for your contribution to this project. I have some doubts: 　１. Is it necessary to isolate all dependencies of this project so that all human cloning repository 　　do not need to install various dependencies separately？ 2. Is it necessary to further describe the operation steps of the demo　to facilitate others to repair 　　and match their own needs？ Looking forward to your early reply and best wish！ Yours, Vickey

    opened by Vickey-ZWQ 1
  • How to deal with .hdfs5 files ?

    How to deal with .hdfs5 files ?

    Hi, astorfi First of all, thanks for your contribution to this project. I have some questions:

    1. Are acoustic features or voice files stored in HDFS files?
    2. How to write the header file of the HDFS file?

    Looking forward to your reply！ Yours, yy

    opened by yy835055664 2
  • Does anyone know the EER of this repo?

    Does anyone know the EER of this repo?

    Hi. I am looking for some speaker verification repo that gives decent EER. I have tried some open source speaker verifications but I haven't been able to find anything that gives below 10 % EER with VoxCeleb1 DB. The speaker verification models availablie online are usually trained with smaller and cleaner DBs and tend not to give reasonable EER on VoxCeleb1. Can anybody please tell me if there is any decent one? I would really appreciate it :D

    opened by hash2430 1
  • Demo video recording link is broken

    Demo video recording link is broken

    The demo video recording link seems to be broken. Would help to view things in action before trying it out locally.

    Link attached to the demo video: https://asciinema.org/a/yfy6FryUAWWMl1vgylrRagMdw

    opened by JudeVJoseph 0
  • where is score_vector.npy

    where is score_vector.npy

    fid = open(os_fspath(file), "rb") FileNotFoundError: [Errno 2] No such file or directory: 'results/SCORES/score_vector.npy' Traceback (most recent call last): File "./code/4-ROC_PR_curve/PlotROC.py", line 73, in

    opened by azuryl 2
Releases(1.1)

  • 1.1(Aug 9, 2017)

    This project is aimed to provide the implementation for Speaker Verification (SR) by using 3D convolutional neural networks following the SR protocol.

    Source code(tar.gz)
    Source code(zip)

  • 1.0(Jul 30, 2017)

    This project is aimed to provide the implementation for Speaker Verification (SR) by using 3D convolutional neural networks following the SR protocol.

    Source code(tar.gz)
    Source code(zip)
Owner
Amirsina Torfi
PhD & Developer working on Deep Learning, Computer Vision & NLP
Amirsina Torfi
GitHub Repository
Simple-System-Convert--C--F - Simple System Convert With Python

Simple-System-Convert--C--F REQUIREMENTS Python version : 3 HOW TO USE Run the c

Jonathan Santos 2 Feb 16, 2022
Mmdet benchmark with python

mmdet_benchmark 本项目是为了研究 mmdet 推断性能瓶颈，并且对其进行优化。 配置与环境 机器配置 CPU：Intel(R) Core(TM) i9-10900K CPU @ 3.70GHz GPU：NVIDIA GeForce RTX 3080 10GB 内存：64G 硬盘：1T

杨培文 (Yang Peiwen) 24 May 21, 2022
Open source Python module for computer vision

About PCV PCV is a pure Python library for computer vision based on the book "Programming Computer Vision with Python" by Jan Erik Solem. More details

Jan Erik Solem 1.9k Jan 06, 2023
🛠️ SLAMcore SLAM Utilities

slamcore_utils Description This repo contains the slamcore-setup-dataset script. It can be used for installing a sample dataset for offline testing an

SLAMcore 7 Aug 04, 2022
KoCLIP: Korean port of OpenAI CLIP, in Flax

KoCLIP This repository contains code for KoCLIP, a Korean port of OpenAI's CLIP. This project was conducted as part of Hugging Face's Flax/JAX communi

Jake Tae 100 Jan 02, 2023
Code for "Finding Regions of Heterogeneity in Decision-Making via Expected Conditional Covariance" at NeurIPS 2021

Finding Regions of Heterogeneity in Decision-Making via Expected Conditional Covariance Justin Lim, Christina X Ji, Michael Oberst, Saul Blecker, Leor

Sontag Lab 3 Feb 03, 2022
Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (ICCV, 2021) (PyTorch) - We released the training code!

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution Kai Zhang, Jingyun Liang, Luc Van Gool, Radu Timofte Computer Vision Lab

Kai Zhang 804 Jan 08, 2023
Drslmarkov - Distributionally Robust Structure Learning for Discrete Pairwise Markov Networks

Distributionally Robust Structure Learning for Discrete Pairwise Markov Networks

1 Nov 24, 2022
Implementation of "Meta-rPPG: Remote Heart Rate Estimation Using a Transductive Meta-Learner"

Meta-rPPG: Remote Heart Rate Estimation Using a Transductive Meta-Learner This repository is the official implementation of Meta-rPPG: Remote Heart Ra

Eugene Lee 137 Dec 13, 2022
SCAN: Learning to Classify Images without Labels, incl. SimCLR. [ECCV 2020]

Learning to Classify Images without Labels This repo contains the Pytorch implementation of our paper: SCAN: Learning to Classify Images without Label

Wouter Van Gansbeke 1.1k Dec 30, 2022
Plenoxels: Radiance Fields without Neural Networks

Plenoxels: Radiance Fields without Neural Networks Alex Yu*, Sara Fridovich-Keil*, Matthew Tancik, Qinhong Chen, Benjamin Recht, Angjoo Kanazawa UC Be

Sara Fridovich-Keil 81 Dec 25, 2022
FcaNet: Frequency Channel Attention Networks

FcaNet: Frequency Channel Attention Networks PyTorch implementation of the paper "FcaNet: Frequency Channel Attention Networks". Simplest usage Models

327 Dec 27, 2022
Contrastive Learning for Many-to-many Multilingual Neural Machine Translation(mCOLT/mRASP2), ACL2021

Contrastive Learning for Many-to-many Multilingual Neural Machine Translation(mCOLT/mRASP2), ACL2021 The code for training mCOLT/mRASP2, a multilingua

104 Jan 01, 2023
A tensorflow model that predicts if the image is of a cat or of a dog.

Quick intro Hello and thank you for your interest in my project! This is the backend part of a two-repo application. The other part can be found here

Tudor Matei 0 Mar 08, 2022
Point Cloud Registration using Representative Overlapping Points.

Point Cloud Registration using Representative Overlapping Points (ROPNet) Abstract 3D point cloud registration is a fundamental task in robotics and c

ZhuLifa 36 Dec 16, 2022
Beancount-mercury - Beancount importer for Mercury Startup Checking

beancount-mercury beancount-mercury provides an Importer for converting CSV expo

Michael Lynch 4 Oct 31, 2022
VLG-Net: Video-Language Graph Matching Networks for Video Grounding

VLG-Net: Video-Language Graph Matching Networks for Video Grounding Introduction Official repository for VLG-Net: Video-Language Graph Matching Networ

Mattia Soldan 25 Dec 04, 2022
Automatic voice-synthetised summaries of latest research papers on arXiv

PaperWhisperer PaperWhisperer is a Python application that keeps you up-to-date with research papers. How? It retrieves the latest articles from arXiv

Valerio Velardo 124 Dec 20, 2022
pytorch implementation of "Contrastive Multiview Coding", "Momentum Contrast for Unsupervised Visual Representation Learning", and "Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination"

Unofficial implementation: MoCo: Momentum Contrast for Unsupervised Visual Representation Learning (Paper) InsDis: Unsupervised Feature Learning via N

Zhiqiang Shen 16 Nov 04, 2020
Face Transformer for Recognition

Face-Transformer This is the code of Face Transformer for Recognition (https://arxiv.org/abs/2103.14803v2). Recently there has been great interests of

Zhong Yaoyao 153 Nov 30, 2022