Systems and methods for multi-speaker neural text-to-speech

US10896669B2 · kind B2 · utility

Assignee

• BAIDU USA LLC · US

Inventors

• Sercan Omer Arik · San Francisco, US
• Gregory Diamos · San Jose, US
• Andrew Gibiansky · Mountain View, US
• John Miller · Redmond, US
• Kainan Peng · Sunnyvale, US
• Wei Ping · Sunnyvale, US
• Jonathan Raiman · Palo Alto, US
• Yanqi Zhou · San Jose, US

Key dates

• Filing date: May 8, 2018
• Grant date: Jan 19, 2021
• Expiry date: Sep 21, 2038

Classification

• Technology area (CPC G): Physics
• CPC primary: G10L25/30
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Described herein are systems and methods for augmenting neural speech synthesis networks with low-dimensional trainable speaker embeddings in order to generate speech from different voices from a single model. As a starting point for multi-speaker experiments, improved single-speaker model embodiments, which may be referred to generally as Deep Voice 2 embodiments, were developed, as well as a post-processing neural vocoder for Tacotron (a neural character-to-spectrogram model). New techniques for multi-speaker speech synthesis were performed for both Deep Voice 2 and Tacotron embodiments on two multi-speaker TTS datasets—showing that neural text-to-speech systems can learn hundreds of unique voices from twenty-five minutes of audio per speaker.