Source Repetition Stimuli

We used a simple generative model to synthesize novel sounds with naturalistic properties. We found that such sounds could be segregated and identified if they occurred more than once across different mixtures, even when the same sounds were impossible to segregate in single mixtures.

Authors:  Josh H. McDermott,  David Wrobleski,  Andrew J. Oxenham
Updated:  2011-01-18
Source:  http://mcdermottlab.mit.edu/source_repetition/index.html
Keywords:  speech,  audition,  sound-sources,  acoustics,  English

Bilingual Aphasia Test (BAT)

The Bilingual Aphasia Test (BAT) was designed to assess each of the languages of a bilingual or multilingual individual with aphasia in an equivalent way.

Authors:  McGill University Neurolinguistics Lab
Updated:  2009-04-30
Source:  https://www.mcgill.ca/linguistics/research/bat
Keywords:  bilinguial,  speech,  aphasia,  language,  linguistics

Cocktail Party Problem

The cocktail party problem is the task of hearing a sound of interest, often a speech signal, in a complex auditory setting.

Authors:  Josh H. McDermott
Updated:  2009-04-30
Source:  http://mcdermottlab.mit.edu/cocktail_examples/index.html
Keywords:  audition,  speech,  noise,  frequency,  English

IEMOCAP: The Interactive Emotional Dyadic Motion Capture Database

An acted, multimodal and multispeaker database containing approximately 12 hours of audiovisual data, including video, speech, motion capture of face, text transcriptions.

Authors:  Carlos Busso,  Murtaza Bulut,  Chi-Chun Lee,  Abe Kazemzadeh,  Emily Mower,  Samuel Kim,  Jeannette N. Chang,  Sungbok Lee,  Shrikanth Narayanan
Updated:  2008-11-09
Source:  https://sail.usc.edu/iemocap/
Keywords:  emotions,  behavior,  speech,  gesture,  motion-capture,  english

British National Corpus

The British National Corpus (BNC) was originally created by Oxford University press in the 1980s - early 1990s, and it contains 100 million words of text from a wide range of genres (e.g. spoken, fiction, magazines, newspapers, and academic).

Authors:  Oxford University Press
Updated:  2007-04-30
Source:  https://www.english-corpora.org/bnc/
Keywords:  English,  linguistics,  corpora,  database,  written-text,  speech

The Dutch Polyphone Corpus

The Dutch Polyphone Corpus contains speech from 5050 speakers from all regions of The Netherlands. Speakers had to read digits and sentences and answer questions over telephone.

Authors:  Els Den Os,  T.I. Boogaart,  Lou Boves,  Esther Klabbers
Updated:  1995-01-30
Source:  https://www.ru.nl/cls/publications/corpora/
Keywords:  language,  speech,  Dutch