Speaker Generation

Abstract

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 attentionbased 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 1 .

Giving Robots a Voice: Human-in-the-Loop Voice Creation and open-ended Labeling

“…Modern TTS models have great factorization abilities [42,82,89], allowing users to independently change text, prosody, and speaker identity (i.e., what and how something is being said by whom). Harnessing such rich latent features [72] not only facilitates the crafting of new voice personae [33,76], but also ensures that these synthesized voices encapsulate the nuances and diversity inherent to human speech.…”