Neural waveform models have demonstrated better performance than conventional
vocoders for statistical parametric speech synthesis. One of the best models, called …

Neural waveform models such as the WaveNet are used in many recent text-to-speech
systems, but the original WaveNet is quite slow in waveform generation because of its …

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 …

Growing research demonstrates that synthetic failure modes imply poor generalization. We
compare commonly used audio-to-audio losses on a synthetic benchmark, measuring the …

Neural sequence-to-sequence text-to-speech synthesis (TTS) can produce high-quality
speech directly from text or simple linguistic features such as phonemes. Unlike traditional …

This paper describes a method to control prosodic features using phonetic and prosodic
symbols as input of attention-based sequence-to-sequence (seq2seq) acoustic modeling …

Neural source-filter (NSF) models are deep neural networks that produce waveforms given
input acoustic features. They use dilated-convolution-based neural filter modules to filter …

When the available data of a target speaker is insufficient to train a high quality speaker-
dependent neural text-to-speech (TTS) system, we can combine data from multiple speakers …

End-to-end text-to-speech (TTS) synthesis is a method that directly converts input text to
output acoustic features using a single network. A recent advance of end-to-end TTS is due …

We have been investigating rakugo speech synthesis as a challenging example of speech
synthesis that entertains audiences. Rakugo is a traditional Japanese form of verbal …