Loading datasets...
License:
CC0-1.0
Steward:
Common Voice
Task: ASR
Release Date: 12/5/2025
Format: MP3
Size: 210.49 MB
Share
A collection of scripted spoken phrases in Kateviri.
Forbidden Usage
It is forbidden to attempt to determine the identity of speakers in the common Voice datasets. It is forbidden to re-host or re-share this 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.
bsh)This datasheet is for version 24.0 of the the Mozilla Common Voice Scripted Speech dataset
for Kateviri (bsh). The dataset contains 6862 clips representing 10.39 hours of recorded
speech (10.24 hours validated) from 14 speakers.
Kateviri is spoken in Afghanistan's province Noristan and in a few regions in Pakistan (Bumburait, Rumbur, Gobor). The number of speakers of Kateviri is around 40,000.
The dataset includes the following distribution of age and gender.
Self-declared gender information, percentage refers to the number of clips annotated with this gender.
| Gender | Percentage |
|---|---|
| Undefined | 100.0% |
Self-declared age information, percentage refers to the number of clips annotated with this age band.
| Age Band | Percentage |
|---|---|
| Undefined | 15.0% |
| Twenties | 57.0% |
| Fifties | 28.0% |
The text corpus consists of transcriptions of spontaneous speech recorded from native Noristani speakers in natural conversations. The content includes everyday topics such as family, culture, daily routines, and local traditions. The dataset contains approximately 3000 to 5000 sentences and was collected primarily from speakers in and Bumburet, Rumbur, and Gabor valleys (Pakistan).
Parso Arabic
ا آ ب پ ت ٹ ث ج چ ڄ ح خ څ د ڈ ر ڑ ز ژ ڗ س ش ݜ ص ض ط ظ ع غ ف ق ک گ ݣ ل م ن ݨ ں ݩ و ہ ھ ٴ ی ے
There follows a randomly selected sample of sentences from the corpus.
شلماچ پشول گوسہ تیو کور اسیش وݜ مع اسیشا پموکچی ائ ایمو تہ کائ نہ ائ
Automatic random samples
اپہ رہ اسہ لوم نعاں کں نئی
گُر داو دیوم نہ بُلہ ای
زوار انگا دے نگݜیے کنہ ستہ ایار
وسݣ عیستہ ارمان یں مع بگورواس گالوم افسوس
بری جِویا نونوا نہ ٹکیہ
Recorded spontaneous conversations
| Domain | Count |
|---|---|
| Undefined | 6862 |
The recordings were collected from native kate speakers, then manually transcribed. Background noise and silence were removed, and irrelevant parts were excluded. The text was checked for accuracy and aligned with the corresponding audio segments
Yes. It is recommended to apply additional noise reduction, normalize audio levels, and perform a manual review of transcriptions for consistency. For advanced use, forced alignment and phonetic labeling can also improve the dataset quality
Nasir Mansoor nasirmansorch@gmail.com
@misc{noristani_dataset_2025,
title ={Noristani Spontaneous Speech Dataset},
author ={Arshad Nasir},
year ={2025},
howpublished ={\url{https://commonvoice.mozilla.org}},
note ={Noristani language corpus for spontaneous speech}}
This dataset was partially funded by the Open Multilingual Speech Fund managed by Mozilla Common Voice.
This dataset is released under the Creative Commons Zero (CC-0) licence. By downloading this data you agree to not determine the identity of speakers in the dataset.