Common Voice Scripted Speech 25.0 - Pashto

License:

CC0-1.0

Steward:

Common Voice

Task: ASR

Release Date: 3/30/2026

Format: MP3

Size: 97.81 GB

Description

A collection of read speech recordings in Pashto (پښتو).

Specifics

Licensing

Creative Commons Zero v1.0 Universal (CC0-1.0)

https://spdx.org/licenses/CC0-1.0.html

Considerations

Restrictions/Special Constraints

None provided.

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.

Processes

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.

Metadata

پښتو — Pashto (`ps`)

This datasheet is for cv-corpus-25.0-2026-03-09 of the Mozilla Common Voice Scripted Speech dataset for Pashto [پښتو - ps]. The dataset contains 4850876 clips representing 5313.64 hours of recorded speech (3041.86 hours validated) from 8239 speakers, recorded from a text corpus of 453,451 sentences.

Language

Accents

Code	Accent	Clips	Speakers
-		482,561 (9.9%)	298 (3.6%)

Demographic information

The dataset includes the following self-declared age and gender distributions. A coverage summary is shown below each table.

Gender

Self-declared gender information. The table shows clip and speaker counts with percentages. Speakers who did not declare a gender are listed as Unspecified. A dash (-) indicates zero.

Code	Gender	Clips	Speakers
male_masculine	Male, masculine	25,078 (0.5%)	36 (0.4%)
female_feminine	Female, feminine	3,041,663 (62.7%)	275 (3.3%)
transgender	Transgender	-	-
non-binary	Non-binary	-	-
do_not_wish_to_say	Prefer not to say	2,687 (0.1%)	6 (0.1%)
-	Unspecified	1,781,448 (36.7%)	8,101 (98.3%)

Gender declared: 3,069,428 of 4,850,876 clips (63.3%), 138 of 8,239 speakers (1.7%)

Age

Self-declared age information. The table shows clip and speaker counts with percentages. Speakers who did not declare an age are listed as Unspecified. A dash (-) indicates zero.

Code	Age	Clips	Speakers
teens	Teens	549,034 (11.3%)	192 (2.3%)
twenties	Twenties	2,958,040 (61.0%)	1,240 (15.1%)
thirties	Thirties	890,010 (18.3%)	536 (6.5%)
fourties	Fourties	144,008 (3.0%)	107 (1.3%)
fifties	Fifties	117,345 (2.4%)	28 (0.3%)
sixties	Sixties	2,603 (0.1%)	8 (0.1%)
seventies	Seventies	92 (0.0%)	2 (0.0%)
eighties	Eighties	-	-
nineties	Nineties	-	-
-	Unspecified	189,744 (3.9%)	7,164 (87.0%)

Age declared: 4,661,132 of 4,850,876 clips (96.1%), 1,075 of 8,239 speakers (13.0%)

Data splits for modelling

Clip buckets

Bucket	Clips
Validated	2,776,943 (57.2%)
Invalidated	177,351 (3.7%)
Other	1,896,582 (39.1%)

Training splits

Split	Clips
Train	218,250 (7.9%)
Dev	15,462 (0.6%)
Test	15,462 (0.6%)

Training split coverage: 249,174 of 2,776,943 validated clips (9.0%)

The dataset contains 2776943 validated, 177351 invalidated, and 1896582 unresolved clips. The average clip duration is 3.943 seconds.

Text corpus

Validated sentences: 334,539

Category	Count
Unvalidated sentences	118,912
Pending sentences	116,020
Rejected sentences	2,892
Reported sentences	6,119

The corpus contains 453,451 sentences: 334,539 validated and 118,912 unvalidated (116,020 pending review, 2,892 rejected), with 6,119 reported for review.

Sample

There follows a randomly selected sample of five sentences from the corpus.

زه له تاسو خوح یم
شکیب ۵۸ قلمونه لري.
مالیزیا د باران او شنو ونو وطن دی.
د کرنې وزارت د حاصلاتو د ودې لپاره د نوو نوښتونو خبر ورکوي.
د میینې لوبې وکړې

Sources

Source	Sentences
Self Citation	126,234 (37.7%)
ځان اقتباس	38,851 (11.6%)
self citation	33,397 (10.0%)
cc0	30,872 (9.2%)
CC0	18,463 (5.5%)
self citation	13,033 (3.9%)
cco	8,925 (2.7%)
فېسبوک	7,180 (2.1%)
خپله	6,926 (2.1%)
Other	50,625 (15.1%)

Text domains

Code	Domain	Clips	Speakers
general	General	567,588 (11.7%)	5,610 (68.1%)
agriculture_food	Agriculture and Food	41,672 (0.9%)	1,688 (20.5%)
automotive_transport	Automotive and Transport	6,086 (0.1%)	666 (8.1%)
finance	Finance	10,454 (0.2%)	693 (8.4%)
service_retail	Service and Retail	25,312 (0.5%)	1,094 (13.3%)
healthcare	Healthcare	88,914 (1.8%)	1,505 (18.3%)
history_law_government	History, Law and Government	164,529 (3.4%)	3,168 (38.5%)
media_entertainment	Media and Entertainment	31,393 (0.6%)	1,428 (17.3%)
nature_environment	Nature and Environment	146,980 (3.0%)	2,983 (36.2%)
news_current_affairs	News and Current Affairs	12,379 (0.3%)	936 (11.4%)
technology_robotics	Technology and Robotics	29,662 (0.6%)	1,215 (14.7%)
language_fundamentals	Language Fundamentals	49,603 (1.0%)	1,597 (19.4%)

Fields

Clips

Each row of a tsv file represents a single audio clip, and contains the following information:

client_id - hashed UUID of a given user
path - relative path of the audio file
text - supposed transcription of the audio
up_votes - number of people who said audio matches the text
down_votes - number of people who said audio does not match text
age - age of the speaker1
gender - gender of the speaker1
accents - accents of the speaker1
variant - variant of the language1
segment - if sentence belongs to a custom dataset segment, it will be listed here
prompt_upvotes - number of upvotes the sentence prompt received
prompt_reports - number of reports the sentence prompt received
is_edited - whether the clip's transcription has been edited

`validated_sentences.tsv`

The validated_sentences.tsv file contains one row per validated sentence in the text corpus:

sentence_id - unique identifier for the sentence
sentence - the sentence text
variant - the variant of the language
sentence_domain - the domain(s) the sentence belongs to
source - the source the sentence was collected from
is_used - whether the sentence is still in circulation for recording
clips_count - number of clips recorded for this sentence

`unvalidated_sentences.tsv`

The unvalidated_sentences.tsv file contains one row per unvalidated sentence in the text corpus:

sentence_id - unique identifier for the sentence
sentence - the sentence text
variant - the variant of the language
sentence_domain - the domain(s) the sentence belongs to
source - the source the sentence was collected from
up_votes - number of upvotes the sentence received
down_votes - number of downvotes the sentence received
status - current status of the sentence (pending or rejected)

Get involved

Community links

Discussions

Contribute

Licence

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.

Footnotes

For a full list of age, gender, and accent options, see the demographics spec. These will only be reported if the speaker opted in to provide that information. ↩ ↩2 ↩3 ↩4