text stringclasses 3
values |
|---|
He holds a magic flute and begins to play. The sweet notes help the train feel better. As he plays his tunes, she feels herself becoming a little braver. |
"" |
The train dances around the boy, cheerful once again. The boy plays all through the night. The train and all the forest animals dance for hours. |
"" |
Omni-StoryBench
Omni-StoryBench is a context-aware omnimodal story generation benchmark.
Each sample provides a current story page and requires generating the next page's image, narration text, and speech utterance.
Dataset Structure
The dataset contains:
data/testset.jsonl: Main benchmark file.images/: Page images.texts/: Page text files.speech/: Generated speech audio files.instruction/: Source-level instruction metadata.
Data Fields
Each JSONL sample includes:
source: Source dataset name.book: Book identifier.prev_key: Current page key.next_key: Target next page key.metadata: Story-level metadata.current_page: Current page image and text.next_page_condition: Condition for next-page generation.next_page: Target next page image, text, and speech paths.speech_metadata_next_page: Speaker and speech attributes.instruction_path: Path to the instruction metadata file.
Intended Use
This dataset is intended for evaluating context-aware omnimodal generation models that jointly generate text, image, and speech for the next page of a story.
Limitations
The dataset is based on publicly available children’s storybooks and may not represent all cultures, narrative styles, visual styles, or speech patterns.
License
This dataset is derived from source storybook datasets. Users should respect the licenses of the original sources.
- Downloads last month
- 2,608