Zobrazeno 1 - 3
of 3
pro vyhledávání: '"Azcarreta, Juan"'
We introduce a novel all neural model for low-latency directional speech extraction. The model uses direction of arrival (DOA) embeddings from a predefined spatial grid, which are transformed and fused into a recurrent neural network based speech ext
Externí odkaz:
http://arxiv.org/abs/2407.04879
Autor:
Ferroni, Giacomo, Turpault, Nicolas, Azcarreta, Juan, Tuveri, Francesco, Serizel, Romain, Bilen, Çagdaş, Krstulović, Sacha
The ranking of sound event detection (SED) systems may be biased by assumptions inherent to evaluation criteria and to the choice of an operating point. This paper compares conventional event-based and segment-based criteria against the Polyphonic So
Externí odkaz:
http://arxiv.org/abs/2010.13648
This work defines a new framework for performance evaluation of polyphonic sound event detection (SED) systems, which overcomes the limitations of the conventional collar-based event decisions, event F-scores and event error rates. The proposed frame
Externí odkaz:
http://arxiv.org/abs/1910.08440