diff --git a/advanced/databases/voxforge/5.json b/advanced/databases/voxforge/5.json
new file mode 100644
index 0000000000000000000000000000000000000000..1365bdf883757485d4b60beb521a0b422b44e103
--- /dev/null
+++ b/advanced/databases/voxforge/5.json
@@ -0,0 +1,41 @@
+{
+ "description": "The VoxForge Database",
+ "root_folder": "/idiap/resource/database/VoxForge/dbase/SpeechCorpus/Trunk/Audio/Main/16kHz_16bit",
+ "protocols": [
+ {
+ "name": "default",
+ "template": "advanced_speaker_recognition/1",
+ "views": {
+ "train": {
+ "view": "Train",
+ "parameters": {}
+ },
+ "dev_templates": {
+ "view": "Templates",
+ "parameters": {
+ "group": "dev"
+ }
+ },
+ "dev_probes": {
+ "view": "Probes",
+ "parameters": {
+ "group": "dev"
+ }
+ },
+ "test_templates": {
+ "view": "Templates",
+ "parameters": {
+ "group": "eval"
+ }
+ },
+ "test_probes": {
+ "view": "Probes",
+ "parameters": {
+ "group": "eval"
+ }
+ }
+ }
+ }
+ ],
+ "schema_version": 2
+}
\ No newline at end of file
diff --git a/advanced/databases/voxforge/5.py b/advanced/databases/voxforge/5.py
new file mode 100644
index 0000000000000000000000000000000000000000..ac3e69855f51457bcc6e937d732f5a08ec81f6a1
--- /dev/null
+++ b/advanced/databases/voxforge/5.py
@@ -0,0 +1,323 @@
+###############################################################################
+# #
+# Copyright (c) 2018 Idiap Research Institute, http://www.idiap.ch/ #
+# Contact: beat.support@idiap.ch #
+# #
+# This file is part of the beat.examples module of the BEAT platform. #
+# #
+# Commercial License Usage #
+# Licensees holding valid commercial BEAT licenses may use this file in #
+# accordance with the terms contained in a written agreement between you #
+# and Idiap. For further information contact tto@idiap.ch #
+# #
+# Alternatively, this file may be used under the terms of the GNU Affero #
+# Public License version 3 as published by the Free Software and appearing #
+# in the file LICENSE.AGPL included in the packaging of this file. #
+# The BEAT platform is distributed in the hope that it will be useful, but #
+# WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY #
+# or FITNESS FOR A PARTICULAR PURPOSE. #
+# #
+# You should have received a copy of the GNU Affero Public License along #
+# with the BEAT platform. If not, see http://www.gnu.org/licenses/. #
+# #
+###############################################################################
+
+import numpy as np
+from collections import namedtuple
+
+from beat.backend.python.database import View
+
+import bob.db.voxforge
+import bob.io.base
+import bob.io.audio
+
+
+#----------------------------------------------------------
+
+
+class Train(View):
+ """Outputs:
+ - speech: "{{ system_user.username }}/array_1d_floats/1"
+ - file_id: "{{ system_user.username }}/text/1"
+ - client_id: "{{ system_user.username }}/text/1"
+
+ One "file_id" is associated with a given "speech".
+ Several "speech" are associated with a given "client_id".
+
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | speech | | speech | | speech | | speech | | speech | | speech |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | file_id | | file_id | | file_id | | file_id | | file_id | | file_id |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ ----------------------------------------------- -----------------------------------------------
+ | client_id | | client_id |
+ ----------------------------------------------- -----------------------------------------------
+ """
+
+ def index(self, root_folder, parameters):
+ Entry = namedtuple('Entry', ['client_id', 'file_id', 'speech'])
+
+ # Open the database and load the objects to provide via the outputs
+ db = bob.db.voxforge.Database()
+
+ objs = sorted(db.objects(groups='world'),
+ key=lambda x: (x.client_id, x.id))
+
+ return [ Entry(x.client_id, x.id, x.make_path(root_folder, '.wav')) for x in objs ]
+
+
+ def get(self, output, index):
+ obj = self.objs[index]
+
+ if output == 'client_id':
+ return {
+ 'text': str(obj.client_id)
+ }
+
+ elif output == 'file_id':
+ return {
+ 'text': str(obj.file_id)
+ }
+
+ elif output == 'speech':
+ audio = bob.io.base.load(obj.speech)
+
+ return {
+ 'value': np.cast['float'](audio[0] * pow(2, 15))
+ }
+
+
+#----------------------------------------------------------
+
+
+class Templates(View):
+ """Outputs:
+ - speech: "{{ system_user.username }}/array_1d_floats/1"
+ - file_id: "{{ system_user.username }}/text/1"
+ - template_id: "{{ system_user.username }}/text/1"
+ - client_id: "{{ system_user.username }}/text/1"
+
+ One "file_id" is associated with a given "speech".
+ Several "speech" are associated with a given "template_id".
+ Several "template_id" are associated with a given "client_id".
+
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | speech | | speech | | speech | | speech | | speech | | speech |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | file_id | | file_id | | file_id | | file_id | | file_id | | file_id |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ ----------------------------------------------- -----------------------------------------------
+ | template_id | | template_id |
+ ----------------------------------------------- -----------------------------------------------
+ -----------------------------------------------------------------------------------------------
+ | client_id |
+ -----------------------------------------------------------------------------------------------
+
+ Note: for this particular database, there is only one "template_id"
+ per "client_id".
+ """
+
+ def index(self, root_folder, parameters):
+ Entry = namedtuple('Entry', ['client_id', 'template_id', 'file_id', 'speech'])
+
+ # Open the database and load the objects to provide via the outputs
+ db = bob.db.voxforge.Database()
+
+ template_ids = db.model_ids(groups=parameters['group'])
+
+ entries = []
+
+ for template_id in template_ids:
+ objs = db.objects(groups=parameters['group'],
+ purposes='enroll',
+ model_ids=[template_id])
+
+ entries.extend([ Entry(x.client_id, template_id, x.id, x.make_path(root_folder, '.wav'))
+ for x in objs ])
+
+ return sorted(entries, key=lambda x: (x.client_id, x.template_id, x.file_id))
+
+
+ def get(self, output, index):
+ obj = self.objs[index]
+
+ if output == 'client_id':
+ return {
+ 'text': str(obj.client_id)
+ }
+
+ elif output == 'template_id':
+ return {
+ 'text': str(obj.template_id)
+ }
+
+ elif output == 'file_id':
+ return {
+ 'text': str(obj.file_id)
+ }
+
+ elif output == 'speech':
+ audio = bob.io.base.load(obj.speech)
+
+ return {
+ 'value': np.cast['float'](audio[0] * pow(2, 15))
+ }
+
+
+#----------------------------------------------------------
+
+
+class Probes(View):
+ """Outputs:
+ - speech: "{{ system_user.username }}/array_1d_floats/1"
+ - file_id: "{{ system_user.username }}/text/1"
+ - probe_id: "{{ system_user.username }}/text/1",
+ - client_id: "{{ system_user.username }}/text/1"
+ - template_ids: "{{ system_user.username }}/array_1d_text/1",
+
+ One "file_id" is associated with a given "speech".
+ One "probe_id" is associated with a given "speech".
+ Several "speech" are associated with a given "client_id".
+ Several "client_id" are associated with a given "template_ids".
+
+ Each probe must be matched against a number of templates defined by a list of
+ client identifiers.
+
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | speech | | speech | | speech | | speech | | speech | | speech |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | file_id | | file_id | | file_id | | file_id | | file_id | | file_id |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ | probe_id | | probe_id | | probe_id | | probe_id | | probe_id | | probe_id |
+ --------------- --------------- --------------- --------------- --------------- ---------------
+ ----------------------------------------------- -----------------------------------------------
+ | client_id | | client_id |
+ ----------------------------------------------- -----------------------------------------------
+ -----------------------------------------------------------------------------------------------
+ | template_ids |
+ -----------------------------------------------------------------------------------------------
+ """
+
+ def index(self, root_folder, parameters):
+ Entry = namedtuple('Entry', ['template_ids', 'client_id', 'probe_id', 'file_id', 'speech'])
+
+ # Open the database and load the objects to provide via the outputs
+ db = bob.db.voxforge.Database()
+
+ template_ids = db.model_ids(groups=parameters['group'])
+
+ template_probes = {}
+ for template_id in template_ids:
+ objs = sorted(db.objects(groups=parameters['group'],
+ purposes='probe',
+ model_ids=[template_id]),
+ key=lambda x: (x.client_id, x.id))
+
+ template_probes[template_id] = [ p.id for p in objs ]
+
+ objs = sorted(db.objects(groups=parameters['group'],
+ purposes='probe'),
+ key=lambda x: (x.client_id, x.id))
+
+ entries = []
+ for obj in objs:
+ templates = [ template_id for template_id in template_ids
+ if obj.id in template_probes[template_id] ]
+ entries.append( (templates, obj) )
+
+ return sorted([ Entry(x[0], x[1].client_id, x[1].id, x[1].id,
+ x[1].make_path(root_folder, '.wav'))
+ for x in entries ],
+ key=lambda x: (len(x.template_ids), x.template_ids, x.client_id, x.file_id))
+
+
+ def get(self, output, index):
+ obj = self.objs[index]
+
+ if output == 'template_ids':
+ return {
+ 'text': [ str(x) for x in obj.template_ids ]
+ }
+
+ elif output == 'client_id':
+ return {
+ 'text': str(obj.client_id)
+ }
+
+ elif output == 'probe_id':
+ return {
+ 'text': str(obj.probe_id)
+ }
+
+ elif output == 'file_id':
+ return {
+ 'text': str(obj.file_id)
+ }
+
+ elif output == 'speech':
+ audio = bob.io.base.load(obj.speech)
+
+ return {
+ 'value': np.cast['float'](audio[0] * pow(2, 15))
+ }
+
+
+#----------------------------------------------------------
+
+
+def setup_tests():
+ # Install a mock load function for the images
+ def mock_load(root_folder):
+ return np.ndarray((1, 512), dtype=np.uint8)
+
+ bob.io.base.load = mock_load
+
+
+#----------------------------------------------------------
+
+
+# Test the behavior of the views (on fake data)
+if __name__ == '__main__':
+
+ setup_tests()
+
+ view = Train()
+ view.objs = view.index(
+ root_folder='',
+ parameters=dict(
+ )
+ )
+ view.get('client_id', 0)
+ view.get('file_id', 0)
+ view.get('speech', 0)
+
+
+ view = Templates()
+ view.objs = view.index(
+ root_folder='',
+ parameters=dict(
+ group='dev',
+ )
+ )
+ view.get('client_id', 0)
+ view.get('template_id', 0)
+ view.get('file_id', 0)
+ view.get('speech', 0)
+
+
+ view = Probes()
+ view.objs = view.index(
+ root_folder='',
+ parameters=dict(
+ group='dev',
+ )
+ )
+ view.get('template_ids', 0)
+ view.get('client_id', 0)
+ view.get('probe_id', 0)
+ view.get('file_id', 0)
+ view.get('speech', 0)
diff --git a/advanced/databases/voxforge/5.rst b/advanced/databases/voxforge/5.rst
new file mode 100644
index 0000000000000000000000000000000000000000..1e5de09a78dffa345abfcb982f04385a7a6a239a
--- /dev/null
+++ b/advanced/databases/voxforge/5.rst
@@ -0,0 +1 @@
+The VoxForge Database
\ No newline at end of file