Loading datasets...
Task: ASR
Release Date: 3/22/2026
Format: MP3
Size: 38.42 MB
Share
A collection of read speech recordings in Norwegian Nynorsk (Norsk (nynorsk)).
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.
nn-NO)This datasheet is for cv-corpus-25.0-2026-03-09 of the Mozilla Common Voice Scripted Speech dataset for Norwegian Nynorsk [Norsk (nynorsk) - nn-NO]. The dataset contains 1558 clips representing 1.92 hours of recorded speech (1.61 hours validated) from 41 speakers, recorded from a text corpus of 5,324 sentences.
Norwegian Nynorsk is a written standard of Norwegian. The standard is the preferred written standard of around 10-15% of the Norwegian population and is largely based on the Norwegian dialects of the West of Norway.
The language code is nn and the locale is nn-NO (Nynorsk as written in Norway). Note that this
locale includes the region code for historical reasons relating to the localisation platform. The dataset
contains Norwegian as spoken anywhere and as written in the Nynorsk standard.
| Code | Accent | Clips | Speakers |
|---|---|---|---|
| - | 1,795 (115.2%) | 15 (36.6%) |
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 | 801 (51.4%) | 13 (31.7%) |
| female_feminine | Female, feminine | 183 (11.7%) | 3 (7.3%) |
| transgender | Transgender | - | - |
| non-binary | Non-binary | - | - |
| do_not_wish_to_say | Prefer not to say | - | - |
| - | Unspecified | 574 (36.8%) | 30 (73.2%) |
Gender declared: 984 of 1,558 clips (63.2%), 11 of 41 speakers (26.8%)
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 | 24 (1.5%) | 1 (2.4%) |
| twenties | Twenties | 382 (24.5%) | 8 (19.5%) |
| thirties | Thirties | 558 (35.8%) | 6 (14.6%) |
| fourties | Fourties | 45 (2.9%) | 2 (4.9%) |
| fifties | Fifties | 10 (0.6%) | 1 (2.4%) |
| sixties | Sixties | - | - |
| seventies | Seventies | - | - |
| eighties | Eighties | - | - |
| nineties | Nineties | - | - |
| - | Unspecified | 539 (34.6%) | 28 (68.3%) |
Age declared: 1,019 of 1,558 clips (65.4%), 13 of 41 speakers (31.7%)
Clip buckets
| Bucket | Clips |
|---|---|
| Validated | 1,307 (83.9%) |
| Invalidated | 61 (3.9%) |
| Other | 190 (12.2%) |
Training splits
| Split | Clips |
|---|---|
| Train | 588 (45.0%) |
| Dev | 322 (24.6%) |
| Test | 392 (30.0%) |
Training split coverage: 1,302 of 1,307 validated clips (99.6%)
The dataset contains 1307 validated, 61 invalidated, and 190 unresolved clips. The average clip duration is 4.441 seconds.
The average length of sentences is 7 tokens (42 characters).
Validated sentences: 5,318
| Category | Count |
|---|---|
| Unvalidated sentences | 6 |
| Pending sentences | - |
| Rejected sentences | 6 |
| Reported sentences | 41 |
The corpus contains 5,324 sentences: 5,318 validated and 6 unvalidated (0 pending review, 6 rejected), with 41 reported for review.
Norwegian is written in the Latin script. The text corpus also contains a limited amount of punctuation.
| Symbol |
|---|
| a |
| å |
| æ |
| b |
| c |
| d |
| e |
| é |
| è |
| f |
| g |
| h |
| i |
| j |
| k |
| l |
| m |
| n |
| o |
| ò |
| ø |
| p |
| q |
| r |
| s |
| t |
| u |
| v |
| w |
| x |
| y |
| z |
| ! |
| - |
| . |
| ? |
There follows a randomly selected sample of five sentences from the corpus.
Det er eit reelt alternativ.
Og i meg sjølv eg kjenner dypter av
Kanskje var det berre i forbifarten.
Kvar eg fekk tak i han?
Eg veit at du kjempar.
Korpus med bokomtalar frå Bokelskere.no (Public domain)
Stortingskorpuset 1.1 (Public domain)
Individual sentences submitted by users through the Mozilla Common Voice interface (Public domain)
| Source | Sentences |
|---|---|
| sentence-collector | 5,054 (95.0%) |
| proverbs http://oaks.nvg.org/norwegian-proverbs.html | 173 (3.3%) |
| bokelskere corpus cc-0 | 55 (1.0%) |
| Other | 36 (0.7%) |
The corpus was initially seeded with sentences from
Book reviews and book discussions (from the web community bokelskere.no)
Transcribed and manually checked discussions from the Norwegian Parliament (Stortinget), from 2017 to 2018
Later, several individual users have contributed sentences in various domains through Mozilla's web interface.
Both sources include sentences in both Norwegian Nynorsk and Norwegian Bokmål.
Norwegian Nynorsk candidate sentences were extracted by both excluding Bokmål-only words and including Nynorsk-only words:
| Only | Words |
|---|---|
| Bokmål | ikke, jeg, en, et, fra, hun, noen, være, mer, ble, mye, bare, boken, kommer, flere, dem |
| Nynorsk | eg, ikkje, ein, me, vere, meir, fleire, berre, ho, eit, blei, vart |
These candidate sentences were then checked by two writers of Nynorsk to ensure that they were indeed written in Nynorsk before inclusion into 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)
Francis M. Tyers
Kevin B. Unhammer
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