Dataset · Hugging Face

iamabdulaziz/karakalpak-stt-news

Name: karakalpak-stt-news
Creator: iamabdulaziz
Keywords: task_categories:automatic-speech-recognition, language:kaa, size_categories:10K<n<100K, format:parquet, modality:audio, modality:text, library:datasets, library:dask, library:polars, library:mlcroissant, region:us, audio, karakalpak, speech

Karakalpak STT News 19,495 Karakalpak news and official-speech recordings, approximately 93.132 hours, with normalized Latin transcripts and fixed train/validation/test splits. Release maintainer: Abdulaziz Yusupaliev. Preparation code, recorded experiments, and publication tools: https://github.com/vvadr/karakalpak-stt-model. Splits Split Clips Hours Train 15,595 74.676 Validation 1,950 9.281 Test 1,950 9.175 Audio is mono 16 kHz PCM16 WAV… See the full description on the dataset page: https://huggingface.co/datasets/iamabdulaziz/karakalpak-stt-news.

View on Hugging FaceReport an issue

Author / organization	iamabdulaziz
Downloads	0 (as reported by Hugging Face)
Likes	0
Created	23 October 2026
Last updated at source	23 October 2026

audiokarakalpakspeech

Embed this entry

<iframe src="https://silytics.si/embed/item/02c8ffd8-3f69-429e-895a-823826a4bed9" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Connected in the knowledge graph

iamabdulaziz/wav2vec2-xls-r-300m-karakalpakModel · used to train

Explore this entry in the graph

More from iamabdulaziz

iamabdulaziz/karakalpak-literary-cleanedDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.