Datasets:

---

aspmirlab

/

BENYO-S2ST-Corpus-1

like 1

This PR enhances the dataset card for BENYO-S2ST-Corpus-1 by adding crucial metadata and improving the content formatting.

Metadata Additions:

• task_categories: Added audio-to-audio for better categorization.
• language: Specified en (English) and yo (Yoruba) as the dataset is bilingual.
• tags: Included relevant tags such as speech-to-speech-translation, low-resource-language, english, yoruba, audio-augmentation, corpus, and africa for enhanced discoverability.
• size_categories: Added 10K<n<100K to indicate the dataset's scale.

Content Improvements:

• Converted the dataset title to a standard Markdown <h1> header.
• Reformatted the existing ArXiv paper link to standard Markdown syntax, including the full paper title.
• Replaced HTML list items (<li>) with Markdown list items (-).
• Cleaned up other HTML tags (<b>, <link>) and special characters for consistent Markdown rendering.
• Improved the formatting of the funding acknowledgment and contact information sections.