MIAO: Multimodal Image-Audio Onomatopoeia Dataset

MIAO is a multimodal dataset consisting of paired sound event clips and onomatopoeic images. It is designed to support research on multimodal correspondence between sounds and visual onomatopoeic expressions. The dataset was authored by KeisukeImoto and was last updated on 2026-05-19.

Use Cases

• Cross-modal retrieval between sound event clips and onomatopoeic images.
• Multimodal representation learning from paired audio and image data.
• Generative modeling tasks for synthesizing sounds from onomatopoeic images or vice versa.

Strengths

• Dataset is explicitly designed for multimodal correspondence research between two distinct modalities.
• The dataset's purpose is clearly stated for tasks like cross-modal retrieval and generative modeling.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

KeisukeImoto on Hugging Face.

Freshness

Last updated 2026-05-19 03:42:38; freshness should be verified.