Timezone: »

Unsupervised Speech Decomposition via Triple Information Bottleneck
Kaizhi Qian · Yang Zhang · Shiyu Chang · Mark Hasegawa-Johnson · David Cox

Tue Jul 14 03:00 PM -- 03:45 PM & Wed Jul 15 04:00 AM -- 04:45 AM (PDT) @

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 difficult and expensive to obtain. In this paper, we propose SpeechSplit, which can blindly decompose speech into its four components by introducing three carefully designed information bottlenecks. SpeechSplit is among the first 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.

Author Information

Kaizhi Qian (University of Illinois at Urbana-Champaign)
Yang Zhang (MIT-IBM Watson AI Lab)
Shiyu Chang (MIT-IBM Watson AI Lab)
Mark Hasegawa-Johnson (University of Illinois)
David Cox (MIT-IBM Watson AI Lab)

More from the Same Authors