Tacotron

lastmod 2019-09-20created 2019-03-063 edits

A text-to-speech synthesis system typically consists of multiple stages, such as a text analysis frontend, an acoustic model and an audio synthesis module. Building these components often requires extensive domain expertise and may contain brittle design choices. In this paper, we present Tacotron, an end-to-end generative text-to-speech model that synthesizes speech directly from characters. Given pairs, the model can be trained completely from scratch with random initialization. We present several key techniques to make the sequence-to-sequence framework perform well for this challenging task. Tacotron achieves a 3.82 subjective 5-scale mean opinion score on US English, outperforming a production parametric system in terms of naturalness. In addition, since Tacotron generates speech at the frame level, it's substantially faster than sample-level autoregressive methods.

Tacotron 2

• Tacotron 2 - Generating Human-like Speech from Text
• Tacotron2 as interpreted by a high school student

Documentation

Tacotron - Towards End-to-End Speech Synthesis

https://arxiv.org/abs/1703.10135

See also

• Speech recognition (Speech-to-Text; STT)
• Speech synthesis (Text-to-Speech; TTS)
• WaveNet
• CycleGan
• DeepVoice

Favorite site

• Github - Tacotron: A TensorFlow implementation of Google's Tacotron speech synthesis with pre-trained model (unofficial)
• Modulabs - (Paper study group) Tacotron : Towards End-to-End Speech Synthesis
• Paper notes by a CS undergrad - Tacotron : Toward End-To-End Speech Synthesis (1)
• Paper notes by a CS undergrad - Tacotron : Toward End-To-End Speech Synthesis (2)