Hausa Speech Dataset

Description

This is a Hausa speech Dataset recorded in three different locations which are: Nile University audio studio, News broadcasting studio, and Quiet rooms at different homes.

Steps to reproduce

The dataset was recorded in three different locations which are Nile University audio studio under the department of mass communication, then some recordings took place in the broadcasting studio while some took place in quiet rooms in different houses. The literatures were read by female and male readers of different ages. The data can further be segmented into unigram, bigram and trigram. Adobe audition was used for recording and segmentation.

Institutions

Categories

Licence