Exploiting Nontrivial Connectivity for Automatic Speech Recognition

Autor:	Paraschiv, Marius, Borgholt, Lasse, Tax, Tycho Max Sylvester, Singh, Marco, Maaløe, Lars
Rok vydání:	2017
Předmět:	Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Nontrivial connectivity has allowed the training of very deep networks by addressing the problem of vanishing gradients and offering a more efficient method of reusing parameters. In this paper we make a comparison between residual networks, densely-connected networks and highway networks on an image classification task. Next, we show that these methodologies can easily be deployed into automatic speech recognition and provide significant improvements to existing models. Comment: Accepted at the ML4Audio workshop at the NIPS 2017
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1711.10271 Zobrazit plný text záznamu View this record from Arxiv