metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: voice
      dtype: string
    - name: duration
      dtype: float64
    - name: transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 103164866.554
      num_examples: 5979
  download_size: 164608881
  dataset_size: 103164866.554
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: apache-2.0
language:
  - crh
task_categories:
  - text-to-speech
size_categories:
  - 1K<n<10K

Open Source Crimean Tatar Text-to-Speech datasets

Community

Discord: https://bit.ly/discord-uds
Speech Recognition: https://t.me/speech_recognition_uk
Speech Synthesis: https://t.me/speech_synthesis_uk

Voices

Male

Abibullah

Quality: high
Duration: 2h + 50m
Audio formats: OPUS
Frequency: 48000 Hz

Arslan

Quality: high
Duration: 40m + 40m
Audio formats: OPUS
Frequency: 48000 Hz

Female

Sevil

Quality: high
Duration: 2h + 29m
Audio formats: OPUS
Frequency: 48000 Hz

Appearance on the web

HF demos:
- https://huggingface.co/spaces/robinhad/qirimtatar-tts

Cite this work

@misc {smoliakov_2025,
    author       = { {Smoliakov} },
    title        = { qirimtatar-tts (Revision c2ceec6) },
    year         = 2025,
    url          = { https://huggingface.co/datasets/Yehor/qirimtatar-tts },
    doi          = { 10.57967/hf/4608 },
    publisher    = { Hugging Face }
}