Loading datasets...
Task: ASR
Release Date: 3/22/2026
Format: MP3
Size: 206.28 MB
Share
A collection of read speech recordings in Northern Hindko (شمالی ہندکو).
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.
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.
hno)This datasheet is for cv-corpus-25.0-2026-03-09 of the Mozilla Common Voice Scripted Speech dataset for Northern Hindko [شمالی ہندکو - hno]. The dataset contains 9407 clips representing 10.47 hours of recorded speech (10.15 hours validated) from 36 speakers, recorded from a text corpus of 2,349 sentences.
Hindko language is spoken mainly in Northwestern region of Pakistan by about eight million people. hno variety of Hindko is used mainly in Hazara region of Khyper Pakhtunkhaw, Kaghan valley & Muzaffarabad division of Pakistan administered Kashmir and in parts of Indian administered Kashmir as well. Oral discourse of Hindko covers a wide range of subjects, including religion, education, poetry, politics, theater etc.
| Code | Accent | Clips | Speakers |
|---|---|---|---|
| - | 2,144 (22.8%) | 1 (2.8%) |
The dataset includes the following self-declared age and gender distributions. A coverage summary is shown below each table.
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 | - | - |
| female_feminine | Female, feminine | 4,553 (48.4%) | 11 (30.6%) |
| transgender | Transgender | - | - |
| non-binary | Non-binary | - | - |
| do_not_wish_to_say | Prefer not to say | - | - |
| - | Unspecified | 4,854 (51.6%) | 31 (86.1%) |
Gender declared: 4,553 of 9,407 clips (48.4%), 5 of 36 speakers (13.9%)
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 | 147 (1.6%) | 2 (5.6%) |
| twenties | Twenties | 6,244 (66.4%) | 21 (58.3%) |
| thirties | Thirties | 2,144 (22.8%) | 1 (2.8%) |
| fourties | Fourties | - | - |
| fifties | Fifties | - | - |
| sixties | Sixties | - | - |
| seventies | Seventies | - | - |
| eighties | Eighties | - | - |
| nineties | Nineties | - | - |
| - | Unspecified | 872 (9.3%) | 24 (66.7%) |
Age declared: 8,535 of 9,407 clips (90.7%), 12 of 36 speakers (33.3%)
Clip buckets
| Bucket | Clips |
|---|---|
| Validated | 9,120 (96.9%) |
| Invalidated | 282 (3.0%) |
| Other | 5 (0.1%) |
Training splits
| Split | Clips |
|---|---|
| Train | 863 (9.5%) |
| Dev | 765 (8.4%) |
| Test | 708 (7.8%) |
Training split coverage: 2,336 of 9,120 validated clips (25.6%)
The dataset contains 9120 validated, 282 invalidated, and 5 unresolved clips. The average clip duration is 4.009 seconds.
The corpus of the langauge is mainly created/written on everyday discourse topics such as geography of the area, langauge, culture, education, sports activities, tourism potential of the area etc. There are more than three thousands sentences in the dataset and word count is around 20 thousand.
Validated sentences: 2,349
| Category | Count |
|---|---|
| Unvalidated sentences | - |
| Pending sentences | - |
| Rejected sentences | - |
| Reported sentences | 21 |
The corpus contains 2,349 sentences: 2,349 validated and 0 unvalidated (0 pending review, 0 rejected), with 21 reported for review.
Perso-Arabic script using a spelling system based on Punjabi is used for the Hindko corpus.
ئ ے ی ء ھ ہ ٷ ؤ و ݩ ں ن م ل گ ک ڨ ق ف غ ع ظ ط ض ص ش س ݬ ز ڑ ر ذ ڈ د ڇ خ ح چ ج ث ٹ ت پ ب ا اّ
There follows a randomly selected sample of five sentences from the corpus.
اے سندھ دریا دے نال نال چلدی ہے۔
کابل دے مسافر اس رستے تو ایندے جلدے ہن۔
ان دے فرش نکیاں یا بڑیاں پتھراں نال بنڑائے جلدے ہن۔
پشور بچ اکثریت مسلماںاں دی ہے۔
اے واقعی دلّا بچ لِہندے اے۔
Written by Muhammad Ilyas (Muzaffarabad), “likhtaan” (Magazine-Abbottabad Mansehra), Hindko Ghar (Channel-Kel Neelum), Hindko-I (Channel-Kutton Neelum), Aprin Hindko (Channel-Abbottabad Mansehra)
| Source | Sentences |
|---|---|
| Self | 2,345 (99.8%) |
| Other | 4 (0.2%) |
General, Agriculture and Food, History, Law and Governmant, Media and Entertainment, Nature and Environment, News and Current Affairs
| Code | Domain | Clips | Speakers |
|---|---|---|---|
| general | General | 2,574 (27.4%) | 34 (94.4%) |
| agriculture_food | Agriculture and Food | 5,816 (61.8%) | 34 (94.4%) |
| automotive_transport | Automotive and Transport | - | - |
| finance | Finance | 3,245 (34.5%) | 34 (94.4%) |
| service_retail | Service and Retail | - | - |
| healthcare | Healthcare | - | - |
| history_law_government | History, Law and Government | - | - |
| media_entertainment | Media and Entertainment | 5,816 (61.8%) | 34 (94.4%) |
| nature_environment | Nature and Environment | - | - |
| news_current_affairs | News and Current Affairs | - | - |
| technology_robotics | Technology and Robotics | - | - |
| language_fundamentals | Language Fundamentals | - | - |
There is limited material available of Hinkdo, the corpus of the langauge is mainly created/written on everyday discourse topics such as geography of the area, langauge, culture, education, sports activities, tourism potential of the area etc. This way, we able to have a more natural form of data/corpus of the language. Language advocates form Abbottabad, Mansehra, Muzaffarabad, Neelum & Jehlum districts are also consulted for validation and their input about Hindko corpus. Social media channels like "Aprin Hinkdo" are also visited to see what is presented about Hinkdo langauge so that most of the aspects of the Hinkdo sould be included in the corpus.
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.tsvThe 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.tsvThe 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)
Muhammad Ilyas ;
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.
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