Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster

Abstract: 

The generalized command response (GCR) model represents intonation as a
superposition of muscle responses to spike command signals. We have previously
shown that the spikes can be predicted by a two-stage system, consisting of a recurrent neural network and a post-processing procedure, but the responses themselves were fixed dictionary atoms. We propose an end-to-end
neural architecture that replaces the dictionary atoms with trainable
second-order recurrent elements analogous to recursive filters. We demonstrate
gradient stability under modest conditions, and show that the system can be
trained by imposing temporal sparsity constraints. Subjective listening tests
demonstrate that the system can synthesize intonation with high naturalness,
comparable to state-of-the-art acoustic models, and retains the physiological
plausibility of the GCR model.

up
0 users have voted:

Paper Details

Authors:
François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner
Submitted On:
10 May 2019 - 11:54am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Marelli Francois, Bastian Schnell, Philip N. Garner
Paper Code:
3481
Document Year:
2019
Cite

Document Files

Presentation Poster

(50)

Subscribe

[1] François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner, "An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4367. Accessed: Nov. 13, 2019.
@article{4367-19,
url = {http://sigport.org/4367},
author = {François Marelli; Bastian Schnell; Hervé Bourlard; Thierry Dutoit; Philip N. Garner },
publisher = {IEEE SigPort},
title = {An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster},
year = {2019} }
TY - EJOUR
T1 - An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster
AU - François Marelli; Bastian Schnell; Hervé Bourlard; Thierry Dutoit; Philip N. Garner
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4367
ER -
François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner. (2019). An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster. IEEE SigPort. http://sigport.org/4367
François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner, 2019. An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster. Available at: http://sigport.org/4367.
François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner. (2019). "An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster." Web.
1. François Marelli, Bastian Schnell, Hervé Bourlard, Thierry Dutoit, Philip N. Garner. An End-to-End Network to Synthesize Intonation using a Generalized Command Response Model - Poster [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4367