Hierarchical Generative Modeling for Controllable Speech Synthesis
This work addresses the challenge of generating customizable speech for applications like voice assistants or audio editing, though it is incremental as it builds on existing VAE and TTS methods.
The paper tackles the problem of controlling latent attributes like speaking style and background noise in speech synthesis, which are rarely annotated, by proposing a hierarchical VAE-based TTS model that achieves controllable synthesis with interpretable attribute groups and fine-grained adjustments.
This paper proposes a neural sequence-to-sequence text-to-speech (TTS) model which can control latent attributes in the generated speech that are rarely annotated in the training data, such as speaking style, accent, background noise, and recording conditions. The model is formulated as a conditional generative model based on the variational autoencoder (VAE) framework, with two levels of hierarchical latent variables. The first level is a categorical variable, which represents attribute groups (e.g. clean/noisy) and provides interpretability. The second level, conditioned on the first, is a multivariate Gaussian variable, which characterizes specific attribute configurations (e.g. noise level, speaking rate) and enables disentangled fine-grained control over these attributes. This amounts to using a Gaussian mixture model (GMM) for the latent distribution. Extensive evaluation demonstrates its ability to control the aforementioned attributes. In particular, we train a high-quality controllable TTS model on real found data, which is capable of inferring speaker and style attributes from a noisy utterance and use it to synthesize clean speech with controllable speaking style.