Text to speech (TTS), or speech synthesis, which aims to synthesize intelligible and natural
speech given text, is a hot research topic in speech, language, and machine learning …

Modern text-to-speech synthesis pipelines typically involve multiple processing stages, each
of which is designed or learnt independently from the rest. In this work, we take on the …

Although neural end-to-end text-to-speech models can synthesize highly natural speech,
there is still room for improvements to its efficiency and naturalness. This paper proposes a …

We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis.
While previous work based on large language models (LLMs) shows impressive …

Abstract Recent Text-to-Speech (TTS) systems trained on reading or acted corpora have
achieved near human-level naturalness. The diversity of human speech, however, often …

Transformer-based text to speech (TTS) model (eg, Transformer TTS~\cite {li2019neural},
FastSpeech~\cite {ren2019fastspeech}) has shown the advantages of training and inference …

This paper presents Non-Attentive Tacotron based on the Tacotron 2 text-to-speech model,
replacing the attention mechanism with an explicit duration predictor. This improves …

With the help of discrete neural audio codecs, large language models (LLM) have
increasingly been recognized as a promising methodology for zero-shot Text-to-Speech …

Despite the ability to produce human-level speech for in-domain text, attention-based end-to-
end text-to-speech (TTS) systems suffer from text alignment failures that increase in …

This paper introduces Parallel Tacotron 2, a non-autoregressive neural text-to-speech
model with a fully differentiable duration model which does not require supervised duration …