Investigation of enhanced Tacotron text-to-speech synthesis systems with self-attention for pitch accent language
This work addresses the challenge of applying end-to-end speech synthesis to Japanese, a difficult language due to character diversity and pitch accents, but it is incremental as it builds on existing Tacotron methods.
The authors tackled end-to-end speech synthesis for Japanese, a pitch accent language, by extending Tacotron with self-attention to capture long-term dependencies, and found that while the proposed systems did not match the quality of top-line pipeline systems, they provided important stepping stones towards this goal.
End-to-end speech synthesis is a promising approach that directly converts raw text to speech. Although it was shown that Tacotron2 outperforms classical pipeline systems with regards to naturalness in English, its applicability to other languages is still unknown. Japanese could be one of the most difficult languages for which to achieve end-to-end speech synthesis, largely due to its character diversity and pitch accents. Therefore, state-of-the-art systems are still based on a traditional pipeline framework that requires a separate text analyzer and duration model. Towards end-to-end Japanese speech synthesis, we extend Tacotron to systems with self-attention to capture long-term dependencies related to pitch accents and compare their audio quality with classical pipeline systems under various conditions to show their pros and cons. In a large-scale listening test, we investigated the impacts of the presence of accentual-type labels, the use of force or predicted alignments, and acoustic features used as local condition parameters of the Wavenet vocoder. Our results reveal that although the proposed systems still do not match the quality of a top-line pipeline system for Japanese, we show important stepping stones towards end-to-end Japanese speech synthesis.