Loading datasets...
Release Date: 6/17/2026
Format: MP3
Size: 3.43 GB
Share
A collection of read speech recordings in Chinese (Hong Kong) (中文（香港）).
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.
zh-HK)This datasheet is for cv-corpus-26.0-2026-06-12 of the Mozilla Common Voice Scripted Speech dataset for Chinese (Hong Kong) [中文（香港） - zh-HK]. The dataset contains 124369 clips representing 143.37 hours of recorded speech (108.54 hours validated) from 3113 speakers, recorded from a text corpus of 20,209 sentences.
| Code | Accent | Clips | Speakers |
|---|---|---|---|
| - | 5,289 (4.3%) | 47 (1.5%) |
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 | 50,176 (40.3%) | 468 (15.0%) |
| female_feminine | Female, feminine | 27,892 (22.4%) | 295 (9.5%) |
| transgender | Transgender | 15 (0.0%) | 1 (0.0%) |
| non-binary | Non-binary | - | - |
| do_not_wish_to_say | Prefer not to say | 99 (0.1%) | 2 (0.1%) |
| - | Unspecified | 46,187 (37.1%) | 2,400 (77.1%) |
Gender declared: 78,182 of 124,369 clips (62.9%), 713 of 3,113 speakers (22.9%)
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 | 2,901 (2.3%) | 57 (1.8%) |
| twenties | Twenties | 40,912 (32.9%) | 392 (12.6%) |
| thirties | Thirties | 13,096 (10.5%) | 217 (7.0%) |
| fourties | Fourties | 15,332 (12.3%) | 111 (3.6%) |
| fifties | Fifties | 3,552 (2.9%) | 28 (0.9%) |
| sixties | Sixties | 809 (0.7%) | 6 (0.2%) |
| seventies | Seventies | 243 (0.2%) | 1 (0.0%) |
| eighties | Eighties | - | - |
| nineties | Nineties | - | - |
| - | Unspecified | 47,524 (38.2%) | 2,365 (76.0%) |
Age declared: 76,845 of 124,369 clips (61.8%), 748 of 3,113 speakers (24.0%)
Clip buckets
| Bucket | Clips |
|---|---|
| Validated | 94,153 (75.7%) |
| Invalidated | 4,713 (3.8%) |
| Other | 25,503 (20.5%) |
Training splits
| Split | Clips |
|---|---|
| Train | 8,452 (9.0%) |
| Dev | 5,604 (6.0%) |
| Test | 5,604 (6.0%) |
Training split coverage: 19,660 of 94,153 validated clips (20.9%)
The dataset contains 94153 validated, 4713 invalidated, and 25503 unresolved clips. The average clip duration is 4.15 seconds.
Validated sentences: 20,186
| Category | Count |
|---|---|
| Unvalidated sentences | 23 |
| Pending sentences | - |
| Rejected sentences | 23 |
| Reported sentences | 814 |
The corpus contains 20,209 sentences: 20,186 validated and 23 unvalidated (0 pending review, 23 rejected), with 814 reported for review.
There follows a randomly selected sample of five sentences from the corpus.
等我除衫先，哎呀戴返頂浴帽先吓
是可憎或是可笑，
放工行過都拉咩，痴Q線
個仔吸毒但係佢無拍低佢，咁咪叫成熟囉
係緊要過千字規範表好多
| Source | Sentences |
|---|---|
| sentence-collector | 19,013 (94.2%) |
| hkgov | 259 (1.3%) |
| Self citation | 252 (1.2%) |
| mtr | 216 (1.1%) |
| Other | 446 (2.2%) |
| Code | Domain | Clips | Speakers |
|---|---|---|---|
| general | General | 1,099 (0.9%) | 40 (1.3%) |
| agriculture_food | Agriculture and Food | - | - |
| automotive_transport | Automotive and Transport | 31 (0.0%) | 15 (0.5%) |
| finance | Finance | 47 (0.0%) | 18 (0.6%) |
| service_retail | Service and Retail | 330 (0.3%) | 34 (1.1%) |
| healthcare | Healthcare | 95 (0.1%) | 25 (0.8%) |
| history_law_government | History, Law and Government | 95 (0.1%) | 29 (0.9%) |
| media_entertainment | Media and Entertainment | 54 (0.0%) | 20 (0.6%) |
| nature_environment | Nature and Environment | 60 (0.0%) | 22 (0.7%) |
| news_current_affairs | News and Current Affairs | 25 (0.0%) | 13 (0.4%) |
| technology_robotics | Technology and Robotics | 63 (0.1%) | 19 (0.6%) |
| language_fundamentals | Language Fundamentals | 92 (0.1%) | 23 (0.7%) |
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)
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