diff --git a/index.xml b/index.xml
index e06b189..d0e2ea2 100644
--- a/index.xml
+++ b/index.xml
@@ -77,7 +77,7 @@
     <package id="pros_cons" name="Pros and Cons" author="Bing Liu" copyright="Copyright (C) 2008 Bing Liu" license="Creative Commons Attribution 4.0 International" licenseurl="http://creativecommons.org/licenses/by/4.0/" webpage="http://www.cs.uic.edu/~liub/FBS/sentiment-analysis.html#datasets" unzip="1" unzipped_size="2921218" size="746276" checksum="c4c7e61fb4d57a2f6c95317194da0f17" subdir="corpora" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/pros_cons.zip" />
     <package id="ptb" name="Penn Treebank" copyright="Copyright (C) 1995 University of Pennsylvania" license="This is a stub for the full Penn Treebank Corpus version 3." unzip="1" unzipped_size="63036" size="6289" checksum="7b633a1b7770279eab00bc1108769c67" subdir="corpora" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/ptb.zip" />
     <package id="punkt" name="Punkt Tokenizer Models" author="Jan Strunk" languages="Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Italian, Malayalam, Norwegian, Polish, Portuguese, Russian, Slovene, Spanish, Swedish, Turkish" unzip="1" unzipped_size="37245719" size="13905355" checksum="8dd1d8760a0976f96e5c262decd75165" subdir="tokenizers" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/tokenizers/punkt.zip" />
-    <package id="punkt_tab" name="Punkt Tokenizer Models" author="Jan Strunk" languages="Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Italian, Malayalam, Norwegian, Polish, Portuguese, Russian, Slovene, Spanish, Swedish, Turkish" unzip="1" unzipped_size="10758927" size="4259017" checksum="350bac9ea63bfa73146fee41b631883c" subdir="tokenizers" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/tokenizers/punkt_tab.zip" />
+    <package id="punkt_tab" name="Punkt Tokenizer Models" author="Jan Strunk" languages="Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Italian, Malayalam, Norwegian, Polish, Portuguese, Russian, Slovene, Spanish, Swedish, Turkish" unzip="1" unzipped_size="10885330" size="4319076" checksum="27ea9ee08fe27fe4484e4c5585cce748" subdir="tokenizers" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/tokenizers/punkt_tab.zip" />
     <package id="qc" name="Experimental Data for Question Classification" author="Xin Li and Dan Roth, UIUC" webpage="http://l2r.cs.uiuc.edu/~cogcomp/Data/QA/QC/" unzip="1" unzipped_size="361090" size="125456" checksum="afd4145ac31cb8d7db715974b9b8b57a" subdir="corpora" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/qc.zip" />
     <package id="reuters" name="The Reuters-21578 benchmark corpus, ApteMod version" webpage="http://kdd.ics.uci.edu/databases/reuters21578/reuters21578.html" license="The copyright for the text of newswire articles and Reuters annotations in the Reuters-21578 collection resides with Reuters Ltd. Reuters Ltd. and Carnegie Group, Inc. have agreed to allow the free distribution of this data *for research purposes only*.  If you publish results based on this data set, please acknowledge its use, refer to the data set by the name 'Reuters-21578, Distribution 1.0', and inform your readers of the current location of the data set." unzip="0" unzipped_size="9073648" size="6378691" checksum="c2acb24d5cccf8035e0fe8d29f440a68" subdir="corpora" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/reuters.zip" />
     <package id="rslp" name="RSLP Stemmer (Removedor de Sufixos da Lingua Portuguesa)" author="Viviane Moreira Orengo (vmorengo@inf.ufrgs.br) and Christian Huyck" languages="Portuguese" unzip="1" unzipped_size="7269" size="3805" checksum="648798996224694251834699fa6e55f7" subdir="stemmers" url="https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/stemmers/rslp.zip" />
diff --git a/packages/tokenizers/punkt_tab.zip b/packages/tokenizers/punkt_tab.zip
index 989f672..5e5ff61 100644
Binary files a/packages/tokenizers/punkt_tab.zip and b/packages/tokenizers/punkt_tab.zip differ