TY - GEN
T1 - Unsupervised speech decomposition via triple information bottleneck
AU - Qian, Kaizhi
AU - Zhang, Yang
AU - Chang, Shiyu
AU - Cox, David
AU - Hasegawa-Johnson, Mark
N1 - Publisher Copyright:
Copyright © 2020 by the Authors. All rights reserved.
PY - 2020
Y1 - 2020
N2 - Speech information can be roughly decomposed into four components: language content, timbre, pitch, and rhythm. Obtaining disentangled representations of these components is useful in many speech analysis and generation applications. Recently, state-of-the-art voice conversion systems have led to speech representations that can disentangle speaker-dependent and independent information. However, these systems can only disentangle timbre, while information about pitch, rhythm and content is still mixed together. Further disentangling the remaining speech components is an under-determined problem in the absence of explicit annotations for each component, which are diffcult and expensive to obtain. In this paper, we propose SPEECHFLOW, which can blindly decompose speech into its four components by introducing three carefully designed information bottlenecks. SPEECHFLOW is among the frst algorithms that can separately perform style transfer on timbre, pitch and rhythm without text labels. Our code is publicly available at https://github.com/auspicious3000/ SpeechSplit.
AB - Speech information can be roughly decomposed into four components: language content, timbre, pitch, and rhythm. Obtaining disentangled representations of these components is useful in many speech analysis and generation applications. Recently, state-of-the-art voice conversion systems have led to speech representations that can disentangle speaker-dependent and independent information. However, these systems can only disentangle timbre, while information about pitch, rhythm and content is still mixed together. Further disentangling the remaining speech components is an under-determined problem in the absence of explicit annotations for each component, which are diffcult and expensive to obtain. In this paper, we propose SPEECHFLOW, which can blindly decompose speech into its four components by introducing three carefully designed information bottlenecks. SPEECHFLOW is among the frst algorithms that can separately perform style transfer on timbre, pitch and rhythm without text labels. Our code is publicly available at https://github.com/auspicious3000/ SpeechSplit.
UR - http://www.scopus.com/inward/record.url?scp=85104000979&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85104000979&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85104000979
T3 - 37th International Conference on Machine Learning, ICML 2020
SP - 7792
EP - 7802
BT - 37th International Conference on Machine Learning, ICML 2020
A2 - Daume, Hal
A2 - Singh, Aarti
PB - International Machine Learning Society (IMLS)
T2 - 37th International Conference on Machine Learning, ICML 2020
Y2 - 13 July 2020 through 18 July 2020
ER -