Multi-speaker Emotional Text-to-speech Synthesizer
This addresses the problem of generating diverse emotional speech for multiple speakers, which is incremental as it builds on existing TTS methods with a structured training strategy.
The paper tackles multi-speaker emotional text-to-speech synthesis by training a model to generate speech for 10 speakers across 7 emotions, using a curriculum learning approach that progresses from single-speaker neutral data to multi-speaker emotional data, resulting in fast learning.
We present a methodology to train our multi-speaker emotional text-to-speech synthesizer that can express speech for 10 speakers' 7 different emotions. All silences from audio samples are removed prior to learning. This results in fast learning by our model. Curriculum learning is applied to train our model efficiently. Our model is first trained with a large single-speaker neutral dataset, and then trained with neutral speech from all speakers. Finally, our model is trained using datasets of emotional speech from all speakers. In each stage, training samples of each speaker-emotion pair have equal probability to appear in mini-batches. Through this procedure, our model can synthesize speech for all targeted speakers and emotions. Our synthesized audio sets are available on our web page.