34 datasets found

French Wikipedia corpus

Asian Scientific Paper Excerpt Corpus (ASPEC)

Swiss SMS corpus dataset

The dataset is used for document-level neural machine translation. It contains 0.23M training sentences, 0.31M development sentences, and 0.21M test sentences.

The Penn Discourse Treebank 2.0 (PDTB 2.0) is a large scale corpus containing 2,312 Wall Street Journal (WSJ) articles.

MS MARCO V1 corpus

A speech corpus of size 7,000 used for training and validation of the FCI module.

The dataset used in this study is the OSCAR corpus, which is a multilingual corpus that is obtained by filtering of the Common Crawl corpus.

A dataset of 2,857 books written by 141 authors, used for pre-training and fine-tuning a language model for author-stylized text generation.

The TIMIT corpus is a widely used benchmark for speech recognition tasks. It contains 3,696 training utterances from 462 speakers, excluding the SA sentences. The core test set...

Sentipers: A sentiment analysis corpus for persian

The Brown corpus is an out-of-domain dataset.

A semantically annotated parallel corpus for English, German, Italian, and Dutch where sentences are aligned with scoped meaning representations in order to capture the...

The Temple University Hospital EEG Seizure corpus (TUSZ) v1.5.2 is a publicly available seizure corpus to date with 5, 612 EDF files, 3, 050 annotated seizures from clinical...

The dataset is a large vocabulary Japanese accent dictionary built using the proposed technique.

The Switchboard corpus is a dataset of speech recordings from a switchboard, which is a device that allows multiple people to speak at the same time.

This dataset comprises more than 150 socio-linguistic interviews with African-American English speakers born between 1891 and 2005.

The Voice Bank speech corpus is a selection of ten British English speakers – both male and female – from the Voice Bank speech corpus, each of which has around 400 clean...

The WSJ corpus is a large vocabulary continuous speech recognition dataset. It contains 36416 sequences, representing around 80 hours of speech.

A new offensive language corpus in Hebrew, manually annotated with a label, targets, topics, and offensive phrases.