RefineGAN: Universally Generating Waveform Better than Ground Truth with Highly Accurate Pitch and Intensity Responses.
Shengyuan XuWenxiao ZhaoJing GuoPublished in: INTERSPEECH (2022)
Keyphrases
- highly accurate
- ground truth
- high quality
- ground truth data
- high accuracy
- capable of producing
- gold standard
- frequency domain
- test images
- database
- accurate models
- quantitative evaluation
- human subjects
- intensity levels
- segmented images
- artificial intelligence
- data sets
- multiresolution
- multiscale
- image gradient
- information systems
- automatically generating
- image segmentation algorithms
- information retrieval
- manually labeled