Common Voice Scripted Speech 25.0 - Svan

License icon

License:

CC0-1.0

Shield icon

Steward:

Common Voice

Task: ASR

Release Date: 3/22/2026

Format: MP3

Size: 353.01 MB

Share

Description

A collection of read speech recordings in Svan (ლუშნუ).

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

ლუშნუ — Svan (sva)

This datasheet is for cv-corpus-25.0-2026-03-09 of the Mozilla Common Voice Scripted Speech dataset for Svan [ლუშნუ - sva]. The dataset contains 10957 clips representing 18.02 hours of recorded speech (15.73 hours validated) from 55 speakers, recorded from a text corpus of 1,549 sentences.

Language

Variants

CodeVariantClipsSpeakers
sva-upperbalლჷჟა̄ბუ5,905 (53.9%)11 (20.0%)
sva-lentekhლჷლტეხუ348 (3.2%)4 (7.3%)
sva-lowerbalლუჩვაბურუ10 (0.1%)2 (3.6%)

Accents

CodeAccentClipsSpeakers
lenjeriლჷჟა̄ბუ, ლენჯა̄̈რ3,693 (33.7%)6 (10.9%)
kalaლჷჟა̄ბუ, კა̄ლ2,467 (22.5%)3 (5.5%)
ushgulლჷჟა̄ბუ, უშგულ2,340 (21.4%)2 (3.6%)
mestiaლჷჟა̄ბუ, მესტია1,492 (13.6%)2 (3.6%)
ipariლჷჟა̄ბუ, იფა̄̈რ1,487 (13.6%)1 (1.8%)
upperbaldaliლჷჟა̄ბუ, და̈ლ1,487 (13.6%)1 (1.8%)
lataliლჷჟა̄ბუ, ლატლი1,487 (13.6%)1 (1.8%)
khelediლჷლტეხუ, ჴელედ348 (3.2%)4 (7.3%)
lentekhiლჷლტეხუ, ლენტხა308 (2.8%)1 (1.8%)
khopuriლჷლტეხუ, ხოფურ308 (2.8%)1 (1.8%)
tskhumariლუჩვაბურუ, ცხჷმა̈რ5 (0.0%)1 (1.8%)
-Other--

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.

CodeGenderClipsSpeakers
male_masculineMale, masculine6 (0.1%)1 (1.8%)
female_feminineFemale, feminine8,751 (79.9%)16 (29.1%)
transgenderTransgender--
non-binaryNon-binary--
do_not_wish_to_sayPrefer not to say--
-Unspecified2,200 (20.1%)48 (87.3%)

Gender declared: 8,757 of 10,957 clips (79.9%), 7 of 55 speakers (12.7%)

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.

CodeAgeClipsSpeakers
teensTeens--
twentiesTwenties1,149 (10.5%)7 (12.7%)
thirtiesThirties2,240 (20.4%)4 (7.3%)
fourtiesFourties2,614 (23.9%)5 (9.1%)
fiftiesFifties2,316 (21.1%)2 (3.6%)
sixtiesSixties1,422 (13.0%)1 (1.8%)
seventiesSeventies457 (4.2%)1 (1.8%)
eightiesEighties--
ninetiesNineties--
-Unspecified759 (6.9%)46 (83.6%)

Age declared: 10,198 of 10,957 clips (93.1%), 9 of 55 speakers (16.4%)

Data splits for modelling

Clip buckets

BucketClips
Validated9,567 (87.3%)
Invalidated1,374 (12.5%)
Other16 (0.1%)

Training splits

SplitClips
Train567 (5.9%)
Dev467 (4.9%)
Test503 (5.3%)

Training split coverage: 1,537 of 9,567 validated clips (16.1%)

The dataset contains 9567 validated, 1374 invalidated, and 16 unresolved clips. The average clip duration is 5.921 seconds.

Text corpus

Validated sentences: 1,538

CategoryCount
Unvalidated sentences11
Pending sentences6
Rejected sentences5
Reported sentences9

The corpus contains 1,549 sentences: 1,538 validated and 11 unvalidated (6 pending review, 5 rejected), with 9 reported for review.

Sample

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

  1. ალი შდჷხჷნდიშ კვა̈ფხ ლი.

  2. ნაცუ̄ლ ლი ლა̄შხი სოფელ.

  3. ა̈გითე ანჴა̈დხ, ეჩქას ესნა̈რ ბაღათჷრშა ჯაჯუ̄ჲ ოხჴერდახ ალჲა̈რს.

  4. ხა̈ლხს ათხიდჷნ გვა̈მიდ.

  5. ლა̄̈თშვ ლახე ზჷ̄სყა̈რ სურუს ხეჴფიხ მა̄რა, ეჯა ჰა̄ვდრი ლა̈ნშა̈ნ ლი.

Sources

SourceSentences
სვანური პროზაული ტექსტები, II, ბალსქვემოური კილო, თბილისი, 1957475 (30.9%)
სვანური ენის ქრესტომათია, თსუ გამომცემლობა, თბილისი, 1978429 (27.9%)
სვანური პროზაული ტექსტები, I, ბალსზემოური კილო, თბილისი, 1939246 (16.0%)
"ვარლამ თოფურიას ""სვანური ენის სახელმძღვანელო"""214 (13.9%)
ზ. ჭუმბურიძე, რ. ქურდაძე, ლ. ნიჟარაძე, სვანური ენა, თბილისი, 200761 (4.0%)
სვანური პროზაული ტექსტები IV, ლაშხური კილო, თბილისი, 197950 (3.3%)
ვარლამ თოფურია, სვანური ენის სახელმძღვანელო, თბილისი, 200819 (1.2%)
Other44 (2.9%)

Text domains

CodeDomainClipsSpeakers
generalGeneral96 (0.9%)14 (25.5%)
agriculture_foodAgriculture and Food108 (1.0%)19 (34.5%)
automotive_transportAutomotive and Transport--
financeFinance--
service_retailService and Retail--
healthcareHealthcare162 (1.5%)22 (40.0%)
history_law_governmentHistory, Law and Government--
media_entertainmentMedia and Entertainment--
nature_environmentNature and Environment--
news_current_affairsNews and Current Affairs--
technology_roboticsTechnology and Robotics--
language_fundamentalsLanguage Fundamentals--

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

Acknowledgements

Funding

This dataset was partially funded by the Open Multilingual Speech Fund managed by Mozilla Common Voice.

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

  1. 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