espnet

/

mixdata_svs_visinger2_spkemb_lang_pretrained

like 1

Follow

ESPnet 316

This Pull Request significantly enhances the model card for SingingSDS: A Singing-Capable Spoken Dialogue System for Conversational Roleplay Applications.

Key improvements include:

• Metadata:
  • Added pipeline_tag: text-to-audio for better discoverability on the Hugging Face Hub.
  • Updated license to mit, aligning with the explicit code license found in the GitHub repository's README.md.
  • Added library_name: espnet, reflecting the primary singing voice synthesis library integral to the system's core functionality and consistent with the existing base_model tag.
• Comprehensive Content:
  • Included a clear overview of the project and its capabilities, derived from the paper abstract and GitHub README.
  • Provided essential links to the paper, GitHub repository, and Hugging Face Space demo, along with a link to the project video playlist.
  • Integrated detailed sections for installation, CLI and web usage, configuration options, project structure, and contributing guidelines, all directly sourced from the GitHub README.
  • Maintained original formatting, including explicit newline characters (\n) in code snippets, as instructed.
  • Included a clear breakdown of the project's various licenses (Code, Character Assets, Model Licenses) for full transparency.
  • Added a BibTeX citation entry for easy academic reference.

This update transforms the model card into a comprehensive resource for users looking to understand and utilize SingingSDS directly from the Hugging Face Hub.