This paper describes a system developed for the GENEA (Generation and
Ev...
We introduce Matcha-TTS, a new encoder-decoder architecture for speedy T...
With read-aloud speech synthesis achieving high naturalness scores, ther...
Spontaneous speech has many affective and pragmatic functions that are
i...
Neural HMMs are a type of neural transducer recently proposed for
sequen...
Neural sequence-to-sequence TTS has achieved significantly better output...