Common Voice Scripted Speech 24.0 - Adyghe

Адыгабзэ — Adyghe (ady)

This datasheet is for version 24.0 of the Mozilla Common Voice Scripted Speech dataset for Adyghe (ady). The dataset contains 63.23 hours of recorded speech (56.76 hours validated) from 137 speakers.

Language

Adyghe (/ˈædɪɡeɪ/ or /ˌɑːdɪˈɡeɪ/, also known as West Circassian) is a Northwest Caucasian language spoken by the western subgroups of Adygheans (Circassians). Identified by the ISO 639-3 code ady, it belongs to the Northwest Caucasian language family. Adyghe is also one of the official languages of the Republic of Adygea, a federal subject of the Russian Federation.

The Adyghean (West Circassians) population is estimated to be between 1,200,000 to 2,000,000 although the number of active speakers is significantly lower. Adygheans are primarily located in the Republic of Adygea, Russia, with additional diaspora populations in Turkey, Jordan, Israel, Syria, the United States, Canada, and various European countries due to historical migrations. However, only about 10% of the Adyghe population is considered literate in the language.

For more information about the global distribution of Adyghe population, see:
The Adyghe (Circassian) Diaspora Today.

Variants

Circassian Language Variants

The following table presents the general language variants of the Adyghe language used in the Common Voice project.

Semtences in the MCV project are provided in Cyrillic and Turkish transliteration to support non-literate users from the Turkish diaspora. With "literacy" we mean being able to read the Adyghean alphabet.

Demographic information

The dataset includes the following distribution of age and gender.

Gender

Self-declared gender information, frequency refers to the number of clips annotated with this gender.

Age

Self-declared age information, frequency refers to the number of clips annotated with this age band.

Data splits for modelling

The official data splits for modelling this language are as follows. Of the validated clips, 31.35% are included in the splits.

Text corpus

The text corpus for the Adyghe language in the Common Voice project consists of a diverse collection of sentences contributed by native speakers. These sentences cover a wide range of topics, including everyday conversations, cultural references, and various domains such as literature, news articles, conversational texts, technology, health, and education. The maximum length of Adyghe sentences in this dataset is limited to 15 words and 140 characters due to restrictions in the Common Voice interface. However, Adyghe sentences can naturally be much longer. In this dataset, the average length has been aimed to be maintained at around 6–8 words and 80–100 characters to reflect the typical structure and complexity of the Adyghe language. The corpus has been curated to ensure a broad representation of vocabulary and grammatical constructs, making it suitable for training robust speech recognition models.

Writing system

The Adyghe language uses the Cyrillic script with some additional letters to represent specific sounds in the language. The alphabet consists of 66 letters, including 33 standard Cyrillic letters and 33 additional letters unique to Adyghe. The writing system is phonetic, meaning that words are generally spelled as they are pronounced. Adyghe is written from left to right.

During the period when there was no Adyghe keyboard option, the recorded text content was written with the Russian keyboard as the closest alternative. The characters in the Adyghe alphabet defined by the Unicode Consortium as "Cyrillic Letter Palocka ('Ӏ': u04C0, utf-8)" and "Cyrillic Letter Small Palocka ('ӏ': u04CF, utf-8)" are not defined in this keyboard model. For this reason, since the early days of computers, when creating digital data in Adyghe or Kabardian, one of the characters "Latin Capital Letter i ('I': u0049, utf-8)", "Latin Small Letter L ('l': u006C, utf-8)" or "Digit One ('1': u0031, utf-8)", albeit rarely "Vertical Line ('|': u007C, utf-8)" was used randomly due to their visual similarities (Nemlioğlu, 2018). This situation emerges as an important factor that negatively affects data quality (Nemlioğlu,2025).

Therefore we allowed the use of only the keyboard layout specially designed for Adyghe language in the Common Voice project. This keyboard layout includes all the letters of the Adyghe alphabet, including the special characters "Cyrillic Letter Palocka ('Ӏ': u04C0, utf-8)" and "Cyrillic Letter Small Palocka ('ӏ': u04CF, utf-8)". The use of this keyboard layout has significantly improved the quality of the text data in Adyghe. Click here to download Adyghe Keyboard for Windows.

Symbol table

Main Alphabet: (Ordered by ascending) а б в г {гу} {гъ} {гъу} д {дж} {дз} {дзу} е ё ж {жъ} {жъу} {жь} з и й к {ку} {къ} {къу} {кӏ} {кӏу} л {лъ} {лӏ} м н о п {пӏ} {пӏу} р с т {тӏ} {тӏу} у ф х {хъ} {хъу} {хь} ц {цу} {цӏ} ч {чъ} {чӏ} ш {шъ} {шъу} {шӏ} {шӏу} щ ъ ы ь э ю я ӏ {ӏу}

Orthographic–Phonetic Table (IPA):

Auxiliary exemplar: The following letters may also be present in the dataset, but are not part of the official Adyghe alphabet. They are used to represent three specific sounds in the Shapsug dialect. They could potentially appear in some dialogues or quoted sentences.

{гь}: Voiced dorso-palatal plosive [ɟ] (e.g. егьэ (jeɟɘ) [Shapsug] = еджэ [Adyghe]: reading) {кь}: Voiceless dorso-palato-velar plosive [kʲ] (e.g. кьэт (kʲɘt) [Shapsug] = чэт [Adyghe]: chicken ) {кӏь}: Glottalized voiceless dorso-palato-velar plosive [kʲʼ] (e.g. кӏьакӏьэ (kʲʼäkʲʼɘ) [Shapsug] = кӏэнкӏэ [Adyghe]: egg)

Other Languages: Some entries in the dataset include words originating from other languages, such as Russian or Turkish. These words have been phonetically transcribed using the Adyghe alphabet, with characters chosen to approximate their original pronunciation as closely as possible.

Sample

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

Sources

All sentences in this dataset were created specifically for the Common Voice project by native Adyghe speakers. The sentences were carefully crafted to cover a wide range of topics and linguistic structures, ensuring a diverse and representative corpus for speech recognition training.

Text domains

Processing

The text data in this dataset has undergone several processing steps to ensure quality and consistency:

Preparation of Sentences:

The sentences prepared by the community were mostly reviewed by Adyghe language experts before being transferred into the system. Prior to transfer, not only spelling errors but also character compatibility (such as corrections of Cyrillic Small Letter Palochka [u04CF] and Cyrillic Letter Palochka [u04C0]) were checked to ensure accuracy.

Normalization:

The text data has been normalized to ensure consistency in formatting and representation. This includes standardizing punctuation, capitalization, and spacing. Special attention was given to the unique characters in the Adyghe alphabet to ensure they are correctly represented.

Validation:

The sentences have been validated by native Adyghe speakers to ensure linguistic accuracy and naturalness. This validation process helps to eliminate any sentences that may be awkward or unnatural in everyday speech.

Filtering:

Sentences that contained inappropriate content, offensive language, or were deemed unsuitable for public use were filtered out during the review process.

Reporting and Corrections:

A system has been established for users to report any errors or issues they encounter in the text data. Reported issues are reviewed by language experts, and necessary corrections are made in subsequent releases of the dataset.

Quality Assurance:

Regular quality assurance checks are conducted to maintain the integrity of the text data. This includes periodic reviews of the sentences to ensure they continue to meet the established standards.

Recommended post-processing

These sentences are provided as-is, and no additional post-processing is required. However, users may choose to perform Unicode normalization (NFC or NFD) depending on their specific use case.

Although the sentences (questions/translation texts) have been written and verified using correct Unicode characters via the ady & kbd keyboard for Windows, it is possible that some text inputs beyond our control may have been submitted due to Common Voice’s open participation model.

After the datasets are published, we re-audit these entries and take the necessary steps to ensure corrections in the next release.

Get involved!

Community links

• If you want to join ady community, start here: https://bit.ly/cv_circassian_start_here

• Common Voice translators on Pontoon

You can find more information about how to participate in the Common Voice Project on the following page:
Community Participation Guidelines

Discussions

Comming soon.

Contribute

We do not advice to contribute to this dataset directly, because of writting system-keyboard issues. Please, read the section "Writing system" above and the following article: https://www.nemerko.org/the-importance-of-adyghe-keyboard-usage-for-artificial-intelligence-training/

If you want to contribute to the Adyghe language, please visit the following links: https://bit.ly/cv_circassian_start_here

A special thanks to all the volunteers who contributed to this dataset.

• Speak

• Write

• Listen

• Review

Acknowledgements

Datasheet authors

M.Ugur Nemlioglu nemerko@nemerko.com Murat Topçu murattopcu67@hotmail.com

Dataset curators

M.Uğur Nemlioğlu nemerko@nemerko.com Saida Abregova Elizaveta Gogunokova Murat Topçu

Advisors

Bülent Özden (Technical Advisor)

Citation guidelines

Funding

This dataset was partially funded by the Open Multilingual Speech Fund.

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.