Deep embeddings with Essentia models

  • Authors
  • Alonso-Jiménez P, Bogdanov D, Serra X
  • UPF authors
  • SERRA, XAVIER; ALONSO JIMENEZ, PABLO; BOGDANOV, DMITRY;
  • Authors of the book
  • AA. VV.
  • Book title
  • Proceedings International Society of Music Information Retrieval Conference (ISMIR 2020)
  • Publisher
  • ISMIR
  • Publication year
  • 2020
  • Pages
  • -
  • Abstract
  • We present the integration of various CNN TensorFlow models developed for different MIR tasks into Essentia. This is a continuation of our previous work, extending the list of supported models and adding new algorithms to facilitate usability. Essentia provides input feature extraction and inference with TensorFlow models in a single C++ pipeline with Python bindings, facilitating the deployment of C++ and Python MIR applications. We assess the new models¿ capabilities to serve as embedding extractors in many downstream classification tasks. All presented models are publicly available on the Essentia website.
  • Complete citation
  • Alonso-Jiménez P, Bogdanov D, Serra X. Deep embeddings with Essentia models. In: AA. VV.. Proceedings International Society of Music Information Retrieval Conference (ISMIR 2020). 1 ed. Montréal: ISMIR; 2020.