This dataset is a curated subset of previously published speech command datasets in Kazakh, Tatar, and Russian. It is intended for use in multilingual speech command recognition and keyword spotting tasks. No data augmentation has been applied. All files are included in their original form as released in the cited works below. This repository simply reorganizes them for convenience and accessibility.

	Languages… See the full description on the dataset page: https://huggingface.co/datasets/artur-muratov/multilingual-speech-commands-3lang-raw.