VISinger2: High-Fidelity End-to-End Singing Voice Synthesis Enhanced by Digital Signal Processing Synthesizer.
Yongmao ZhangHeyang XueHanzhao LiLei XieTingwei GuoRuixiong ZhangCaixia GongPublished in: INTERSPEECH (2023)
Keyphrases
- end to end
- high fidelity
- digital signal processing
- signal processing
- text to speech
- data flow
- low power
- real time
- image processing
- computer vision and image processing
- high quality
- congestion control
- medical image compression
- high resolution
- image quality
- low cost
- artificial intelligence
- scalable video
- power consumption
- human operators
- image compression
- pattern recognition
- neural network