Fine-grained robust prosody transfer for single-speaker neural text-to-speech
This work addresses a specific bottleneck in text-to-speech synthesis for applications requiring speaker variability, but it is incremental as it builds on existing prosody transfer methods.
The paper tackles the problem of prosody transfer in neural text-to-speech when using a single-speaker dataset, which is not robust to unseen speakers, by decoupling reference signal alignment and using phoneme-level features with a variational auto-encoder. The result is a more stable system that reliably transfers prosody from unseen speakers, as validated by objective and subjective tests.
We present a neural text-to-speech system for fine-grained prosody transfer from one speaker to another. Conventional approaches for end-to-end prosody transfer typically use either fixed-dimensional or variable-length prosody embedding via a secondary attention to encode the reference signal. However, when trained on a single-speaker dataset, the conventional prosody transfer systems are not robust enough to speaker variability, especially in the case of a reference signal coming from an unseen speaker. Therefore, we propose decoupling of the reference signal alignment from the overall system. For this purpose, we pre-compute phoneme-level time stamps and use them to aggregate prosodic features per phoneme, injecting them into a sequence-to-sequence text-to-speech system. We incorporate a variational auto-encoder to further enhance the latent representation of prosody embeddings. We show that our proposed approach is significantly more stable and achieves reliable prosody transplantation from an unseen speaker. We also propose a solution to the use case in which the transcription of the reference signal is absent. We evaluate all our proposed methods using both objective and subjective listening tests.