CommonLanguage

CommonLanguage Dataset 

This dataset is composed of speech recordings from of languages that were carefully selected from the CommonVoice database. The total duration of audio recordings is 45.1 hours (i.e., 1 hour of material for each language). The dataset is already balanced and split into train, dev (validation) and test sets. The dataset has been extracted from CommonVoice to train easily language-id systems.

The baselines for language-id are available in the SpeechBrain toolkit (see recipes/CommonLanguage):
https://github.com/speechbrain/speechbrain

Statistics of CommonLanguage:

| Name                              | Train  | Dev    | Test  |
|:---------------------------------:|:------:|:------:|:-----:|
| **# of utterances**               | 177552 | 47104  | 47704 |
| **# unique speakers**             | 11189  | 1297   | 1322  |
| **Total duration, hr**            | 30.04  | 7.53   | 7.53  |
| **Min duration, sec**             | 0.86   | 0.98   | 0.89  |
| **Mean duration, sec**            | 4.87   | 4.61   | 4.55  |
| **Max duration, sec**             | 21.72  | 105.67 | 29.83 |
| **Duration per language, min**    | ~40    | ~10    | ~10   |

## List of languages:
* Arabic
* Basque
* Breton
* Catalan
* Chinese_China
* Chinese_Hongkong
* Chinese_Taiwan
* Chuvash
* Czech
* Dhivehi
* Dutch
* English
* Esperanto
* Estonian
* French
* Frisian
* Georgian
* German
* Greek
* Hakha_Chin
* Indonesian
* Interlingua
* Italian
* Japanese
* Kabyle
* Kinyarwanda
* Kyrgyz
* Latvian
* Maltese
* Mangolian
* Persian
* Polish
* Portuguese
* Romanian
* Romansh_Sursilvan
* Russian
* Sakha
* Slovenian
* Spanish
* Swedish
* Tamil
* Tatar
* Turkish
* Ukranian
* Welsh

## Other information
In addition to the language label, the datapoints have `age`, `gender` and `utterance transcription` associated with each utterance.