Common Voice Scripted Speech 25.0 - Japanese

License:

CC0-1.0

Steward:

Common Voice

Task: ASR

Release Date: 3/23/2026

Format: MP3

Size: 14.34 GB

Description

A collection of read speech recordings in Japanese (日本語).

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

日本語 — Japanese (`ja`)

This datasheet is for cv-corpus-25.0-2026-03-09 of the Mozilla Common Voice Scripted Speech dataset for Japanese [日本語 - ja]. The dataset contains 584432 clips representing 725.11 hours of recorded speech (371.92 hours validated) from 7813 speakers, recorded from a text corpus of 51,207 sentences.

Language

Accents

Code	Accent	Clips	Speakers
-		53,100 (9.1%)	431 (5.5%)

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	156,678 (26.8%)	1,744 (22.3%)
female_feminine	Female, feminine	228,052 (39.0%)	2,261 (28.9%)
transgender	Transgender	203 (0.0%)	2 (0.0%)
non-binary	Non-binary	143 (0.0%)	4 (0.1%)
do_not_wish_to_say	Prefer not to say	5,158 (0.9%)	73 (0.9%)
-	Unspecified	194,068 (33.2%)	4,935 (63.2%)

Gender declared: 390,364 of 584,432 clips (66.8%), 2,878 of 7,813 speakers (36.8%)

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	47,749 (8.2%)	636 (8.1%)
twenties	Twenties	318,967 (54.6%)	4,004 (51.2%)
thirties	Thirties	40,270 (6.9%)	282 (3.6%)
fourties	Fourties	48,396 (8.3%)	265 (3.4%)
fifties	Fifties	25,213 (4.3%)	215 (2.8%)
sixties	Sixties	5,435 (0.9%)	53 (0.7%)
seventies	Seventies	578 (0.1%)	7 (0.1%)
eighties	Eighties	-	-
nineties	Nineties	110 (0.0%)	2 (0.0%)
-	Unspecified	97,714 (16.7%)	3,932 (50.3%)

Age declared: 486,718 of 584,432 clips (83.3%), 3,881 of 7,813 speakers (49.7%)

Data splits for modelling

Clip buckets

Bucket	Clips
Validated	299,767 (51.3%)
Invalidated	55,101 (9.4%)
Other	229,564 (39.3%)

Training splits

Split	Clips
Train	19,695 (6.6%)
Dev	9,019 (3.0%)
Test	9,019 (3.0%)

Training split coverage: 37,733 of 299,767 validated clips (12.6%)

The dataset contains 299767 validated, 55101 invalidated, and 229564 unresolved clips. The average clip duration is 4.467 seconds.

Text corpus

Validated sentences: 43,942

Category	Count
Unvalidated sentences	7,265
Pending sentences	4,718
Rejected sentences	2,547
Reported sentences	791

The corpus contains 51,207 sentences: 43,942 validated and 7,265 unvalidated (4,718 pending review, 2,547 rejected), with 791 reported for review.

Sample

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

今の自民党政権に戻るとの意味では全くない。中道改革勢力を結集し、私たちが政権を担える政治を目指すということだ
金融機関に連絡する義務がある。
本人が謝るのが筋だろうに
そして、これが空いている時間です。
I want to eat an apple!

Sources

Source	Sentences
sentence-collector	11,437 (26.1%)
covost2-en_ja	6,713 (15.3%)
自己引用	3,171 (7.2%)
Self Citation	2,798 (6.4%)
JSUT	2,176 (5.0%)
Aozora Bunko	1,994 (4.5%)
yumie-text-1	1,579 (3.6%)
https://slib.net/50151	1,229 (2.8%)
?????	786 (1.8%)
Other	12,010 (27.4%)

Text domains

Code	Domain	Clips	Speakers
general	General	24,569 (4.2%)	2,845 (36.4%)
agriculture_food	Agriculture and Food	9,173 (1.6%)	2,319 (29.7%)
automotive_transport	Automotive and Transport	2,152 (0.4%)	1,323 (16.9%)
finance	Finance	2,214 (0.4%)	1,249 (16.0%)
service_retail	Service and Retail	490 (0.1%)	417 (5.3%)
healthcare	Healthcare	2,647 (0.5%)	1,512 (19.4%)
history_law_government	History, Law and Government	4,744 (0.8%)	1,630 (20.9%)
media_entertainment	Media and Entertainment	5,068 (0.9%)	1,789 (22.9%)
nature_environment	Nature and Environment	2,761 (0.5%)	1,530 (19.6%)
news_current_affairs	News and Current Affairs	1,068 (0.2%)	825 (10.6%)
technology_robotics	Technology and Robotics	674 (0.1%)	411 (5.3%)
language_fundamentals	Language Fundamentals	3,849 (0.7%)	1,657 (21.2%)

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