Quantum Vision theory transforms speech spectrograms and MFCCs into information waves, enabling QV-CNN and QV-ViT models to reach up to 94.57% accuracy on the ASVspoof deepfake detection dataset.
Natural tts synthesis by conditioning wavenet 24 on mel spectrogram predictions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Quantum Vision Theory Applied to Audio Classification for Deepfake Speech Detection
Quantum Vision theory transforms speech spectrograms and MFCCs into information waves, enabling QV-CNN and QV-ViT models to reach up to 94.57% accuracy on the ASVspoof deepfake detection dataset.