AudioHate-DB: An audio dataset for English hate speech detection in Bengali accent
Arif Faisal, Syeda Sumaia Sultana, Suprove Chandra Sarkar, Shabnom Mustary · 2026
Read AudioHate-DB: An audio dataset for English hate speech detection in Bengali accent by Arif Faisal, Syeda Sumaia Sultana, Suprove Chandra Sarkar, Shabnom Mustary (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13627115.txt.
Description
Faisal and colleagues introduce in 2026 AudioHate-DB, a speech dataset for detecting hate in spoken English delivered with a Bengali accent. Written hate speech resources are plentiful, but accented speech from South Asia is barely covered. That gap holds back both detection systems and speech recognition.
The dataset holds 3,018 phone-recorded audio samples built from a controlled inventory of 1,006 sentences, split almost evenly between hateful and neutral content. Hateful sentences each carry at least one term from a list of 23 slurs and profanities sorted into six categories. Speakers of different genders from several regions recorded in ordinary indoor and outdoor settings, and both raw and noise-cleaned versions are supplied. Baseline tests with four standard models topped out with a recurrent network reaching accuracy just under eighty-four percent on this dataset.
Catalog record
| CREATOR | Arif Faisal, Syeda Sumaia Sultana, Suprove Chandra Sarkar, Shabnom Mustary |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |