Loading datasets...
License:
CC0-1.0
Steward:
Common Voice
Task: ASR
Release Date: 9/15/2025
Format: MP3
Size: 205.72 MB
Share
A collection of scripted spoken phrases in Nawdm.
Restrictions/Special Constraints
You agree that you will not re-host or re-share this dataset
Forbidden Usage
You agree not to attempt to determine the identity of speakers in the Common Voice dataset
Intended Use
This dataset is intended to be used for training and evaluating automatic speech recognition (ASR) models. It may also be used for applications relating to computer-aided language learning (CALL) and language or heritage revitalisation.
nmz)This datasheet is for the Mozilla Common Voice Scripted Speech dataset
for Nawdem (nmz). The dataset contains 11.63 hours of recorded speech (collected, ongoing validation) from more than 35 recording endpoints, covering age ranges from twenties to seventies.
Nawdm (also known as Nawdem, Losso, Losu, Naoudem) is a Gur language spoken in northern Togo and southern Ghana. It belongs to the Niger‑Congo family, under the Oti‑Volta subgroup. The language is used in daily communication, religious texts, poetry, oral tradition, and common expressions.
Nawdm is a tonal language, with at least two level tones:
High tone: marked with an acute accent (á)
Low tone: marked with a grave accent (à)
Tone marking is not always used, typically reserved for pronouns, religious or formal texts.
Uppercase letters:
A B D E Ɛ F G Gw Gb H Ĥ I J K Kw Kp L M N Ny Ŋ Ŋm O Ɔ R S T U V W Y
Lowercase letters:
a b d e ɛ f g gw gb h ɦ i j k kw kp l m n ny ŋ ŋm o ɔ r s t u v w y
The dataset has been compiled from:
Religious texts and translations
Dictionaries of Nawdm
Common expressions collected from everyday life and fieldwork
To get more information about this dataset and find out how you can contribute, please take a look at the following links.
Common Voice Nawdm (nmz) page (link to be activated when available)
Justin Bakoubolo
Guedela, PhD
Justin Bakoubolo — supervised the compilation and coordination of the dataset.
Justin Bakoubolo <justin.bakoubolo@umbaji.org>
This dataset was made possible by the Open Multilingual Speech Fund.
This dataset is released under the Creative Commons Zero (CC‑0) licence. By downloading this data, you agree not to identify individual speakers in the dataset.