Audio feature information generation method and device, equipment and storage medium

• Main Authors: SUN JIANQING, LIANG JIA'EN, YU ZUDA
• Format: Patent
• Language: Chinese; English
• Published: 06.12.2022
• Subjects: ACOUSTICS; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

The invention discloses an audio feature information generation method and device, equipment and a storage medium. The method comprises the following steps: receiving text information to be subjected to speech synthesis and a selected sound type; predicting acoustic feature information based on the text information; obtaining a standardized parameter corresponding to the sound type; performing reduction processing on the acoustic feature information by using the standardized parameters; generating audio feature information corresponding to the sound type according to the restored acoustic feature information; the audio feature information is used for generating a voice signal corresponding to the sound type. In the invention, a plurality of voices share the same group of predictors to predict the acoustic feature information, the predictors output normalized acoustic features, and the acoustic features can be inversely normalized to the original space through the standardized parameters corresponding to the s