Loading datasets...
Task: ASR
Release Date: 6/17/2026
Format: MP3
Size: 347.02 MB
Share
A collection of read speech recordings in Balti (بلتی).
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.
bft)This datasheet is for cv-corpus-26.0-2026-06-12 of the Mozilla Common Voice Scripted Speech dataset for Balti [بلتی - bft]. The dataset contains 10547 clips representing 17.36 hours of recorded speech (16.56 hours validated) from 155 speakers, recorded from a text corpus of 7,980 sentences.
Balti is known as the lingua franca in Baltistan, the Northern Areas of Pakistan. It is spoken in the four districts of Baltistan Division: Skardu, Ghanche, Shigar, Kharmang and some parts of Round District of Gilgit Baltistan. It is spoken by Eight Lac people.
| Code | Accent | Clips | Speakers |
|---|---|---|---|
| - | 45 (0.4%) | 5 (3.2%) |
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 | 1,327 (12.6%) | 12 (7.7%) |
| transgender | Transgender | - | - |
| non-binary | Non-binary | - | - |
| do_not_wish_to_say | Prefer not to say | 1,032 (9.8%) | 1 (0.6%) |
| - | Unspecified | 8,188 (77.6%) | 148 (95.5%) |
Gender declared: 2,359 of 10,547 clips (22.4%), 7 of 155 speakers (4.5%)
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 | 101 (1.0%) | 1 (0.6%) |
| twenties | Twenties | 3,863 (36.6%) | 24 (15.5%) |
| thirties | Thirties | 3,703 (35.1%) | 3 (1.9%) |
| fourties | Fourties | 1 (0.0%) | 1 (0.6%) |
| fifties | Fifties | - | - |
| sixties | Sixties | - | - |
| seventies | Seventies | - | - |
| eighties | Eighties | - | - |
| nineties | Nineties | - | - |
| - | Unspecified | 2,879 (27.3%) | 143 (92.3%) |
Age declared: 7,668 of 10,547 clips (72.7%), 12 of 155 speakers (7.7%)
Clip buckets
| Bucket | Clips |
|---|---|
| Validated | 10,062 (95.4%) |
| Invalidated | 464 (4.4%) |
| Other | 21 (0.2%) |
Training splits
| Split | Clips |
|---|---|
| Train | 2,211 (22.0%) |
| Dev | 1,833 (18.2%) |
| Test | 1,808 (18.0%) |
Training split coverage: 5,852 of 10,062 validated clips (58.2%)
The dataset contains 10062 validated, 464 invalidated, and 21 unresolved clips. The average clip duration is 5.928 seconds.
The corpus have been collected from Balti books written by local Balti poets and writers.
Validated sentences: 7,968
| Category | Count |
|---|---|
| Unvalidated sentences | 12 |
| Pending sentences | 12 |
| Rejected sentences | - |
| Reported sentences | 37 |
The corpus contains 7,980 sentences: 7,968 validated and 12 unvalidated (12 pending review, 0 rejected), with 37 reported for review.
Perso-Arabic script has been used as the writing system of Balti language.
اب پ ت ٹ ث ج ڃ چ ڇ ح خ د ڈ ذ ر ڑ زڗ ژ س ش ݜ ص ض ط ظ ع غ ف ق ک گ کٔ ل م ن نٚ ݨ و ہ ء ی ے
There follows a randomly selected sample of five sentences from the corpus.
زیربا چِکنا، لے اَتی رِنموشے، مو حرام زادان٘و دُوکپا یودپو مُشکل یود
اےسی ایو بی رَگی قبضِنگ لقپو تونگ
اے جق لہ کھونگ ینگ پولو لہ سونگس
خیر کھونگ دیکھہ زین تُھون٘ین دُوکس
مِندوغی قسم گنگمہ نہ اومی ہرکونگ نہ شرابی ہرکونگ نہ گنگمہ یودسُوک
Ehsan Ali Danish; novels and poetry Younus Jones
| Source | Sentences |
|---|---|
| Balti Zdrung | 6,434 (80.7%) |
| Self | 1,093 (13.7%) |
| Balti Novel | 366 (4.6%) |
| Other | 75 (0.9%) |
General
Collected Balti books in soft form and extracted more than ten thousand sentences from the books with the prior consents of the respective writers.
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
sentence - the sentence to be read aloud
sentence_id - unique identifier for the sentence
sentence_domain - domain classification(s) of the sentence
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
locale - locale code of the language
segment - if sentence belongs to a custom dataset segment, it will be listed here
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)
Common Voice Community
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