Audio generator and method for generating audio signal and for training audio generator

Techniques for generating an audio signal and training an audio generator are disclosed. An audio generator (10) may generate an audio signal (16) from target data (12) representing the audio signal (16) and an input signal (14). Target data (12) is derived from text. The audio generator comprises: a first processing block (40, 50, 50a-50h) receiving first data (15, 59a) derived from an input signal (14) and outputting first output data (69); and a second processing block (45) receiving the first output data (69) or data derived from the first output data (69) as second data. The first processing block (50) comprises: a condition set of learnable layers (71, 72, 73) configured to process the target data (12) to obtain condition feature parameters (74, 75); and a pattern element (77) configured to apply the conditional characteristic parameter (74, 75) to the first data (15, 59a) or the normalized first data (59, 76 '). 公开了用于生成音频信号和训练音频发生器的技术。音频发生器(10)可以从表示音频信号(16)的目标数据(12)和输入信号(14)生成音频信号(16)。目标数据(12)从文本得出。音频发