Unsupervised Speech Decomposition via Triple Information Bottleneck
This addresses the challenge of under-determined speech decomposition for applications in speech analysis and generation, offering a novel unsupervised method.
The paper tackles the problem of disentangling speech into four components (language content, timbre, pitch, and rhythm) without explicit annotations, proposing SpeechSplit, which achieves unsupervised decomposition and enables style transfer on timbre, pitch, and rhythm.
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.