metame-ai/awesome-audio-plaza

Curated collection aggregating papers, projects, datasets, and toolkits across nine audio domains—including audio encoding, speech translation, emotion recognition, audio separation, and voice omni-models. Content is automatically sourced from arXiv, Hugging Face Papers, GitHub trending, Papers with Code, and social media, then organized into category-specific markdown docs with surveys, implementations, and evaluation resources. Each domain includes dedicated sections for research papers, open-source projects, commercial products, and benchmark datasets to support both academic research and production deployment.

411 stars.