Speaker Generation Article Swipe

PDF

Related Concepts

Computer science Task (project management) Speech recognition Speaker diarisation Similarity (geometry) Speaker recognition Embedding Artificial intelligence Natural language processing Sampling (signal processing) Perception Image (mathematics) Psychology Filter (signal processing) Management Neuroscience Computer vision Economics

Daisy Stanton , Matt Shannon , Soroosh Mariooryad , RJ Skerry-Ryan , Eric Battenberg , Tom Bagby , David Kao ·

YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2111.05095 · OA: W4306247551

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.