Speech to Speech Synthesis for Voice Impersonation
This addresses voice impersonation, a domain-specific problem, with incremental improvements over existing methods.
The paper tackles speech-to-speech style transfer for voice impersonation by proposing the Speech to Speech Synthesis Network (STSSN), which fuses speech recognition and synthesis to generate realistic audio samples, showing more convincing results than a generative adversarial model in benchmarks.
Numerous models have shown great success in the fields of speech recognition as well as speech synthesis, but models for speech to speech processing have not been heavily explored. We propose Speech to Speech Synthesis Network (STSSN), a model based on current state of the art systems that fuses the two disciplines in order to perform effective speech to speech style transfer for the purpose of voice impersonation. We show that our proposed model is quite powerful, and succeeds in generating realistic audio samples despite a number of drawbacks in its capacity. We benchmark our proposed model by comparing it with a generative adversarial model which accomplishes a similar task, and show that ours produces more convincing results.