flashdata-dev's picture
Fix license and checksum file links
cb007c0 verified
|
Raw History Blame Contribute Delete
6.14 kB
metadata
language:
  - en
license: cc-by-4.0
size_categories:
  - n<1K
pretty_name: YouTube-Commons — FlashData Transcript Sample
tags:
  - video-metadata
  - youtube
  - flashdata
  - custom-video-dataset
configs:
  - config_name: default
    data_files:
      - split: sample
        path: sample.parquet

YouTube-Commons — FlashData Transcript Sample

12 English transcripts with creator credits and source metadata, covering practical tutorials, agriculture, culture and learning.

12 records · Full transcripts and creator metadata · Parquet + JSONL · No video files

Request a custom video dataset · Explore the YouTube Download API

Use this sample to inspect a source manifest, test a loader and prepare a scoped collection request. For playable MP4 examples, see FlashData's video dataset Collection.

Preview

Video ID Example metadata
tvMTjEfnMUw Creator @Jamesmatthews on how to film a documentary #filmmaking #videoediting #artlist
hhrHtRx5aIY StckRateScenarios#2
UDZyBEsdq4o The Crosh Cuirn: A Manx tradition for May Day Eve
Vsd0n93nzns Coes Sorghum

Load the sample

from datasets import load_dataset

sample = load_dataset("flashdata-dev/youtube-commons-transcript-sample", split="sample")
print(len(sample))  # 12
print(sample[0]["text"])

Install datasets first if needed. Direct files: Parquet, JSONL, provenance, checksums, quality report.

Prepare your source list

Download prepare_delivery.py and sample.jsonl into the same folder, then run:

python prepare_delivery.py sample.jsonl --output delivery-manifest.jsonl --report normalization-report.json

The helper accepts local CSV or JSONL, normalizes YouTube IDs, deduplicates sources and accounts for every input row. It makes no network requests and does not download media. The output is a source manifest for scoping a delivery, not an API request payload.

For a custom collection, provide the source list or topic, target languages, desired resolution, clip boundaries, caption needs and storage destination. FlashData can scope available media delivery and metadata packaging against the requested permissions and acceptance criteria. See the YouTube Download API documentation for the current API contract.

Fields

Field Meaning
video_id, video_link, source_video_url Original ID and link; source_video_url is an alias
title, channel, channel_id, date Original title, creator attribution and upload date
text Complete, unchanged upstream transcript
license Original upstream video-license statement
original_language, transcription_language, language_id_method Preserved upstream language metadata
word_count, character_count Original upstream counts; not recomputed
topic, source_row_index FlashData topic and original zero-based row position

Source and selection

Upstream: https://huggingface.co/datasets/PleIAs/YouTube-Commons. Exact source revision and record positions are in provenance.json.

Twelve explicitly selected rows from the first 4,096 rows of cctube_0.parquet, with English original and transcription language metadata. The chosen row indices are frozen in provenance.json. This is a small convenience sample, not a balanced corpus.

Changes made by FlashData: Preserved every original field and the full original transcript text. Added a manually assigned topic, source_video_url alias and zero-based source_row_index. No transcript correction, translation, alignment or new ASR was performed.

License and attribution

The upstream transcript/metadata release is labeled CC BY 4.0. Preserve the individual creator credits when redistributing records. The original license string is retained as reported by the source; this package does not independently certify each uploader's rights or assign a new license version to the underlying videos. Transcripts can contain recognition errors and source claims; topic labels are FlashData's additions.

YouTube-Commons by Pleias and its contributors. Each source creator is credited by channel, channel_id, title and video_link in every row and in ATTRIBUTION.md.

See LICENSE, ATTRIBUTION.md and LICENSE-CODE. FlashData selected and packaged this sample independently. No affiliation or endorsement by the upstream projects or creators is implied.

Validation and limits

All 12 records load locally through Hugging Face Datasets; JSONL and Parquet match exactly. IDs are unique and syntactically valid. Every package file has a SHA-256 checksum. No source video was downloaded or checked for current availability, and no transcript or annotation accuracy score is claimed. Upstream content, availability and rights may differ across records. This small sample is intended for inspecting formats and workflows.

References