PAZHVAK
Here we introduce PAZHVAK the first public dataset for Persian voice recognition (words). The dataset contains 88535 audio files of 4018 common words in Persian. For each word, 18 to 40 audio files have been recorded and put in a folder labeled by a number. The corresponding Persion word of each folder can be found in Labels File. The voice files were recorded by 61 senior bachelor students of computer engineering (38 male and 23 female) at University of Hormozgan in 2024.
| .Folder No | File Size | Link |
|---|---|---|
| 1 to 400 | 461 MB | 1-400 |
| 401 to 800 | 512 MB | 401-800 |
| 801 to 1200 | 425 MB | 801-1200 |
| 1201 to 1600 | 345 MB | 1201-1600 |
| 1601 to 2000 | 418 MB | 1601-2000 |
| 2001 to 2400 | 361 MB | 2001-2400 |
| 2401 to 2800 | 307 MB | 2401-2800 |
| 2801 to 3200 | 357 MB | 2801-3200 |
| 3201 to 3600 | 341 MB | 3201-3600 |
| 3601 to 4018 | 340 MB | 3601-4018 |
For citaiton:
