GALE Word Aligned and Tagged Text:  Arabic/Chinese and English parallel word aligned corpora LDC developed for the DARPA GALE program

GALE Arabic-English Parallel Aligned Treebank -- Broadcast News Part 2 [1]: over 141,000 tokens of word aligned Arabic and English parallel text with treebank annotations from Arabic source broadcast programming collected by LDC in 2007 and 2008

GALE Chinese-English Word Alignment and Tagging – Broadcast Training Part 1 [2]: over 179,000 tokens from Chinese source broadcast programming collected by LDC in 2005-2007 (***This is currently on the page)

Arabic Speech Database: King Saud University Arabic Speech Database [3]: 590 hours of recorded Arabic speech from 269 male and female speakers in quiet and noisy environments with speaker metadata

Open Machine Translation (OpenMT) Evaluation Series: LDC’s ongoing effort to support research in machine translation (MT) technologies that translate all forms of text between human languages

NIST 2012 Open Machine Translation (OpenMT) Progress Test Five Language Source [4]: 20 files that include the evaluation sets (source data and human reference translations), DTD, scoring software, and evaluation plan for the OpenMT 2012 test for Arabic, Chinese, Dari, Farsi, and Korean to English on a parallel data set

CALLFRIEND Farsi Second Edition: the release of the entire collection of telephone conversations among native Farsi speakers -- originally recorded in 1995 and 1996 – and their transcripts

CALLFRIEND Farsi Second Edition Speech [5]:  over 42 hours of telephone conversations (100 recordings) among native Farsi speakers in the continental United States making a single call to a family member or friend living in the United States  

CALLFRIEND Farsi Second Edition Transcripts [6]:  the complete set of transcripts in three formats (Romanized, Arabic-script and XML) 

Maninkakan Lexicon: part of an LDC ongoing project to build an electronic dictionary of four Mandekan languages -- Mawukakan, Maninkakan, Bambara and Jula

Maninkakan Lexicon [7]: 5834 Maninkakan entries presented as a Maninkakan-English lexicon and a Maninkakan French-lexicon using a Latin-based transcription system in Toolbox and XML formats

The ARRAU (Anaphora Resolution and Underspecification) Project: annotations of multi-genre English texts for anaphoric relations with information about agreement and explicit respresentation of multiple antecedents for anaphoric expresssions and discourse antecedents

The ARRAU Corpus of Anaphoric Information [8]: task-oriented dialogues, narratives and newswire annotated for semantic roles, morphosyntactic agreement, grammatical function, semantic type and genericity