NEURAL SOURCE-FILTER-BASED WAVEFORM MODEL FOR STATISTICAL PARAMETRIC SPEECH SYNTHESIS

Authors: Mingyang Zhang, Xin Wang, Fuming Fang, Haizhou Li, Junichi Yamagishi
preprint paper,

1: YOU HAVE ALL THE ADVANTAGE

Source:
Target:
	TTS_standalone	VC_standalone	Hybrid_TTS	Hybrid_VC	Hybrid_TTS&VC
Outputs:

2: AH IT WAS SWEET IN MY EARS

Source:
Target:
	TTS_standalone	VC_standalone	Hybrid_TTS	Hybrid_VC	Hybrid_TTS&VC
Outputs:

3: THIS WAS WHEN THE EXPLOSION OCCURRED

Source:
Target:
	TTS_standalone	VC_standalone	Hybrid_TTS	Hybrid_VC	Hybrid_TTS&VC
Outputs:

4: THE GABRIEL VOICE OF THE SAMURAI RANG OUT

Source:
Target:
	TTS_standalone	VC_standalone	Hybrid_TTS	Hybrid_VC	Hybrid_TTS&VC
Outputs:

5: THE HISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT

Source:
Target:
	TTS_standalone	VC_standalone	Hybrid_TTS	Hybrid_VC	Hybrid_TTS&VC
Outputs:

Acknowledgement
These synthetic speech samples were constructed using the CMU Arctic database. The CMU_ARCTIC databases were constructed at the Language Technologies Institute at Carnegie Mellon University. See http://festvox.org/cmu_arctic/ for more details.