Speaker Generation
This addresses the need for generating diverse, synthetic voices in applications like virtual assistants or entertainment, though it appears incremental as it builds on existing text-to-speech models.
The paper tackles the problem of synthesizing speech in nonexistent human-sounding voices, called speaker generation, and presents TacoSpawn, a system that performs competitively at this task by learning a distribution over a speaker embedding space to sample novel speakers.
This work explores the task of synthesizing speech in nonexistent human-sounding voices. We call this task "speaker generation", and present TacoSpawn, a system that performs competitively at this task. TacoSpawn is a recurrent attention-based text-to-speech model that learns a distribution over a speaker embedding space, which enables sampling of novel and diverse speakers. Our method is easy to implement, and does not require transfer learning from speaker ID systems. We present objective and subjective metrics for evaluating performance on this task, and demonstrate that our proposed objective metrics correlate with human perception of speaker similarity. Audio samples are available on our demo page.