Datasets:

freococo
/

eastern_poe_karen_asr

license: cc0-1.0
language:
  - kjp
pretty_name: Eastern Poe Karen ASR
tags:
  - automatic-speech-recognition
  - audio
  - karen
  - burmese
  - myanmar
  - webdataset
  - public-domain
task_categories:
  - automatic-speech-recognition
  - audio-to-audio
  - audio-classification
language_creators:
  - found
source_datasets:
  - original

This is the first public Eastern Poe Karen language ASR dataset in AI history.

Eastern Poe Karen ASR

This dataset contains audio recordings and aligned metadata in the Eastern Poe Karen language (a regional variety of Eastern Pwo, ISO 639-3: pwo), a Karenic language spoken primarily in Mon State and Kayin State in southeastern Myanmar. While linguistically described as Eastern Pwo Karen, the community and this project prefer the term Poe as a community-endorsed spelling.

All audio segments in this dataset were sourced from publicly available news broadcasts published by PVTV (People’s Voice Television), the official ethnic news program of the National Unity Government (NUG) of Myanmar. These broadcasts include multilingual content and promote information access in ethnic languages.

The dataset includes over 11.58 hours of segmented and labeled audio, prepared in WebDataset format, with paired .audio and .json files suitable for training automatic speech recognition (ASR) systems.

Acknowledgments

Special thanks to:

PVTV (People’s Voice Television) for producing and distributing public-interest ethnic news
National Unity Government (NUG) for supporting ethnic language broadcasts
Language volunteers and researchers working on inclusive speech technology for Myanmar

Dataset Structure & Format

This dataset follows the WebDataset format. Each training sample consists of two paired files inside a tar archive:

XXXX.audio — the audio chunk (in MP3 format)
XXXX.json — the corresponding metadata (UTF-8 JSON)

🟢 Minimum chunk duration: 2.04 sec
🔴 Maximum chunk duration: 15.05 sec

Each .json file contains the following fields:

{
  "file_name": "XXXX.audio",
  "video_id": "YouTubeVideoID",
  "title": "Original broadcast title from PVTV",
  "url": "https://www.youtube.com/watch?v=YouTubeVideoID",
  "duration": 13.24
}

Usage Example

You can load and stream this dataset using the Hugging Face datasets library:

    from datasets import load_dataset

    dataset = load_dataset(
        "freococo/eastern_poe_karen_asr",
        split="train",
        streaming=True
    )

    for sample in dataset:
        print(sample["audio"])        # Audio object
        print(sample["file_name"])    # Chunk filename
        print(sample["duration"])     # Duration in seconds
        print(sample["title"])        # Broadcast title
        print(sample["url"])          # YouTube source URL

Each sample includes:

audio: the audio chunk (in MP3 format, stored as .audio)
file_name: the filename of the chunk
title: broadcast title in Burmese
url: direct YouTube source link
video_id: YouTube video ID
duration: duration of the audio in seconds

Known Limitations

This dataset was segmented automatically from PVTV news broadcasts using pause-based or fixed-length chunking. As such:

No transcriptions are available.
Some chunks may contain music, jingles, or non-speech background audio.
There is no speaker annotation or speech-vs-noise filtering.
Audio conditions vary widely, reflecting real-world broadcast variability.

Despite these limitations, this dataset offers a high-coverage, community-oriented resource for speech technology in Eastern Poe Karen.

Licensing & Use

All content is released under the Creative Commons Zero (CC0 1.0 Universal) public domain dedication.

You are free to:

- Use and remix the data
- Train models (open or commercial)
- Publish research or applications

Please respect the context and dignity of the original speakers and news publishers.

📚 Citation

Freococo (2025).
Eastern Poe Karen ASR
https://huggingface.co/datasets/freococo/eastern_poe_karen_asr
Dataset compiled from PVTV ethnic news broadcasts in the Eastern Poe Karen language.
Released under CC0 1.0 (Public Domain).