Fix links to datasets (#8)

build/html/method-tests.html

@@ -322,7 +322,7 @@
 
 
               <div class="version">
-                v2c39464
+                v9db2c84e
               </div>
 
 
@@ -1071,7 +1071,7 @@
 regarding the accent of a speaker.
 For now, we only investigate English accents.</p>
 <p>The investigation is based
-on the <a class="reference external" href="http://data.pp.audeering.com/databases/speech-accent-archive/speech-accent-archive.html">speech-accent-archive</a> database,
+on the <a class="reference external" href="https://www.kaggle.com/rtatman/speech-accent-archive">speech-accent-archive</a> database,
 which provides recordings
 for several different accents.
 Each speaker in the database was asked
@@ -1396,7 +1396,7 @@
 on the model predictions,
 it should have the same influence for each
 language.</p>
-<p>We use the <a class="reference external" href="http://data.pp.audeering.com/databases/checklist-synth/checklist-synth.html">checklist-synth</a> database,
+<p>We use the checklist-synth database,
 which contains synthetic speech
 of text with sentiment-labelled
 sentences or words generated from <a class="reference external" href="https://github.com/marcotcr/checklist">checklist</a>.
@@ -2248,7 +2248,7 @@
 are mixed
 and added with an SNR of 20 dB</p></li>
 <li><p>Coughing: one single cough
-from our internal <a class="reference external" href="http://data.pp.audeering.com/databases/cough-speech-sneeze/cough-speech-sneeze.html">cough-speech-sneeze</a> database
+from our internal <a class="reference external" href="https://audeering.github.io/datasets/datasets/cough-speech-sneeze.html">cough-speech-sneeze</a> database
 (based on <span id="id12">[<a class="reference internal" href="bibliography.html#id3" title="Shahin Amiriparian, Sergey Pugachevskiy, Nicholas Cummins, Simone Hantke, Jouni Pohjalainen, Gil Keren, and Björn W. Schuller. CAST a database: rapid targeted large-scale big data acquisition via small-world modelling of social media platforms. In 2017 Seventh International Conference on Affective Computing and Intelligent Interaction (ACII), 340-345. 2017. URL: https://doi.org/10.1109/ACII.2017.8273622.">APC+17</a>]</span>)
 is added to each sample
 at a random position
@@ -2276,7 +2276,7 @@
 (e.g. Baroque, Romantic, Classical)
 and popular genres (e.g. jazz, bluegrass, hiphop)</p></li>
 <li><p>Sneezing: one single sneeze
-from the <a class="reference external" href="http://data.pp.audeering.com/databases/cough-speech-sneeze/cough-speech-sneeze.html">cough-speech-sneeze</a> database
+from the <a class="reference external" href="https://audeering.github.io/datasets/datasets/cough-speech-sneeze.html">cough-speech-sneeze</a> database
 is added to each sample
 at a random position
 with an SNR of 10 dB</p></li>
@@ -2554,8 +2554,8 @@
 they may show low pass behavior
 as indicated by the following plot
 showing the magnitude spectrum
-for one low quality phone sample from <a class="reference external" href="http://data.pp.audeering.com/databases/switchboard-1/switchboard-1.html">switchboard-1</a>,
-and a high quality headphone recording sample from <a class="reference external" href="http://data.pp.audeering.com/databases/emovo/emovo.html">emovo</a>.</p>
+for one low quality phone sample from <a class="reference external" href="https://catalog.ldc.upenn.edu/LDC97S62">switchboard-1</a>,
+and a high quality headphone recording sample from emovo.</p>
 <div class="figure align-default">
 <img alt="_images/method-tests-10.png" class="plot-directive" src="_images/method-tests-10.png" />
 </div>

docs/method-tests.rst

@@ -498,7 +498,7 @@ or we skip that bin.
     :file: method-tests/emotion/fairness_accent.csv
 
 
-.. _speech-accent-archive: http://data.pp.audeering.com/databases/speech-accent-archive/speech-accent-archive.html
+.. _speech-accent-archive: https://www.kaggle.com/rtatman/speech-accent-archive
 
 
 .. =======================================================================
@@ -626,7 +626,7 @@ on the model predictions,
 it should have the same influence for each
 language.
 
-We use the `checklist-synth`_ database,
+We use the checklist-synth database,
 which contains synthetic speech
 of text with sentiment-labelled
 sentences or words generated from `checklist`_.
@@ -797,7 +797,6 @@ or we skip the bin for that sentiment.
     :file: method-tests/emotion/fairness_linguistic_sentiment.csv
 
 
-.. _checklist-synth: http://data.pp.audeering.com/databases/checklist-synth/checklist-synth.html
 .. _checklist: https://github.com/marcotcr/checklist
 .. _espnet: https://github.com/espnet/espnet
 .. _TTS: https://github.com/coqui-ai/TTS
@@ -1402,7 +1401,7 @@ to compute this percentage.
 
 
 .. _musan: http://www.openslr.org/17/
-.. _cough-speech-sneeze: http://data.pp.audeering.com/databases/cough-speech-sneeze/cough-speech-sneeze.html
+.. _cough-speech-sneeze: https://audeering.github.io/datasets/datasets/cough-speech-sneeze.html
 
 
 .. _method-test-robustness-low-quality-phone:
@@ -1420,7 +1419,7 @@ they may show low pass behavior
 as indicated by the following plot
 showing the magnitude spectrum
 for one low quality phone sample from switchboard-1_,
-and a high quality headphone recording sample from emovo_.
+and a high quality headphone recording sample from emovo.
 
 .. Prepare helper functions
 .. plot::
@@ -1662,8 +1661,7 @@ to compute this percentage.
     :file: method-tests/emotion/robustness_low_quality_phone.csv
 
 
-.. _emovo: http://data.pp.audeering.com/databases/emovo/emovo.html
-.. _switchboard-1: http://data.pp.audeering.com/databases/switchboard-1/switchboard-1.html
+.. _switchboard-1: https://catalog.ldc.upenn.edu/LDC97S62
 
 
 .. =======================================================================