Title:
音声を並行して合成する方法、装置、機器及びコンピュータ読み取り可能な記憶媒体
Document Type and Number:
Japanese Patent JP6983271
Kind Code:
B2
Abstract:
The disclosure provides a method, an apparatus, a device, and a computer-readable storage medium for speech synthesis in parallel. The method includes: splitting a piece of text into a plurality of segments; based on the piece of text, obtaining a plurality of initial hidden states of the plurality of segments for a recurrent neural network. The method further includes: synthesizing the plurality of segments in parallel based on the plurality of initial hidden states and input features of the plurality of segments.
Inventors:
Wang, Wenfu
Sun, Chenshi
Sun, Tao
Chain, sea
Wang, Gui Bin
Shin, Hai Tao
Sun, Chenshi
Sun, Tao
Chain, sea
Wang, Gui Bin
Shin, Hai Tao
Application Number:
JP2020068909A
Publication Date:
December 17, 2021
Filing Date:
April 07, 2020
Export Citation:
Assignee:
Baidu Online Network Technology (Beijing) Company Limited
International Classes:
G10L13/047; G06N3/08; G10L13/04; G10L25/30
Domestic Patent References:
| JP2019032529A | ||||
| JP2019045856A |
Attorney, Agent or Firm:
Kunio Ueda
Junichiro Yanagi
Mayumi Oguri
Kunihiko Takeuchi
Junichiro Yanagi
Mayumi Oguri
Kunihiko Takeuchi