Cross-speaker style transfer (CSST) in text-to-speech (TTS) synthesis aims at transferring a speaking style to the synthesised speech in a target speaker's voice. Most previous CSST approaches rely on expensive high-quality data carrying desired speaking style during training and require a reference utterance to obtain speaking style descriptors as conditioning on the generation of a new sentence. This work presents Referee, a robust reference-free CSST approach for expressive TTS, which fully leverages low-quality data to learn speaking styles from text. Referee is built by cascading a text-to-style (T2S) model with a style-to-wave (S2W) model. Phonetic PosteriorGram (PPG), phoneme-level pitch and energy contours are adopted as fine-grained speaking style descriptors, which are predicted from text using the T2S model. A novel pretrain-refinement method is adopted to learn a robust T2S model by only using readily accessible low-quality data. The S2W model is trained with high-quality target data, which is adopted to effectively aggregate style descriptors and generate high-fidelity speech in the target speaker's voice. Experimental results are presented, showing that Referee outperforms a global-style-token (GST)-based baseline approach in CSST.
翻译:在文本到语音(TTS)的合成中,跨语音风格传输(CSST)旨在将发言风格转换成以目标发言者声音合成的语音。前的CSST方法大多依赖高价高质量数据,在培训期间带有理想的语音风格,需要参考语句表达,以获得语音风格描述器作为生成新句子的附加条件。这份工作展示了Referee,这是用于表达 TTS的一种强有力的无参考的CSST方法,它充分利用低质量数据从文本中学习语音风格。被选者是通过一种带有样式到波模式模型的文本到式(T2S)模式构建的。音频式PoicesterGram(PG)、电话级音级音调和能源配置器被采纳为精准的语音风格描述器,这是用 T2S 模型从文本预测的。采用了一种新型的、基于前置力的数据来学习稳健的 T2S2S模式。S2W 模式由一种具有风格到风格的文本模式模式模式模式,它能以高质量的GS-Sdeal-Sdealal-Spealal-Spealal-Speal-Speal 演示结果,它以展示了高质量的G-Speal-Speal-Speal-Speal-Speal-Speal-Speal-Speal-S-S-S