KSE-RESEARCH-Group/ukr-dialects-audio-dataset

--- language: - uk license: cc-by-4.0 task_categories: - automatic-speech-recognition tags: - ukrainian - dialect - speech - audio configs: - config_name: default data_files: - split: train path: data/train-* - split: validation path: data/validation-* - split: test path: data/test-* dataset_info: features: - name: audio dtype: audio: sampling_rate: 16000 - name: text dtype: string - name: transcribed_text dtype: string - name: sentence_id dtype: int64 - name: segment dtype: string - name: duration dtype: float64 - name: alignment_score dtype: float64 - name: word_count dtype: int64 - name: part dtype: string splits: - name: train num_bytes: 7405819611.322855 num_examples: 27520 - name: validation num_bytes: 881914989.92422 num_examples: 3347 - name: test num_bytes: 923835590.6766155 num_examples: 3435 download_size: 8859666414 dataset_size: 9211570191.92369 --- # Ukrainian Dialects Audio Dataset Merged Ukrainian dialect speech dataset combining 5 speaker datasets, with train/validation/test splits. ## Dataset Description This dataset contains audio recordings of Ukrainian dialect speech, merged from the following source datasets: - [NaUKMA-Audio-Dataset](https://huggingface.co/datasets/KSE-RESEARCH-Group/NaUKMA-Audio-Dataset) - [Ivanna-Stefiuk-Audio-Dataset](https://huggingface.co/datasets/KSE-RESEARCH-Group/Ivanna-Stefiuk-Audio-Dataset) - [Larysa-Irodenko-Audio-Dataset](https://huggingface.co/datasets/KSE-RESEARCH-Group/Larysa-Irodenko-Audio-Dataset) - [Hutsulendia-Audio-Dataset](https://huggingface.co/datasets/KSE-RESEARCH-Group/Hutsulendia-Audio-Dataset) - [Dido-Yvanchyk-Audio-Dataset-v2](https://huggingface.co/datasets/KSE-RESEARCH-Group/Dido-Yvanchyk-Audio-Dataset-v2) ## Dataset Structure - **train**: 27,675 samples - **validation**: 3,365 samples - **test**: 3,451 samples - **Total**: 34,491 samples ## Features - `audio`: Audio file (16 kHz, mono) - `text`: Original text - `transcribed_text`: Transcribed text - `sentence_id`: Sentence identifier - `segment`: Segment identifier - `duration`: Audio duration in seconds - `alignment_score`: Alignment quality score - `word_count`: Number of words - `part`: Part identifier ## Usage ```python from datasets import load_dataset dataset = load_dataset("KSE-RESEARCH-Group/ukr-dialects-audio-dataset") ``` ## License CC-BY-4.0