A Russian dialect speech dataset for dialectological and speech technology research

Russian Dialect Speech Dataset was created for research purposes in both linguistic research and speech processing tasks. The dataset contains over 110,000 audio files, more than 125 hours of audio. The data cover a wide range of macrodialects and dialect groups, offering broad and diverse representation of spoken variation. The recordings were collected under natural field conditions. The audio fragments are saved in .wav format and converted to a sampling rate of 16 kHz, which is the standard for speech recognition tasks.Dataset is provided free for scientific research purposes. Usage of the dataset is based on the terms and conditions of the data use agreement.