Result Details
Task: Deepfake Audio Detection
                Analysis Date: 2025-09-02 12:19:19.248039
                
                    
                        Information:
Codec: mp3 | Bitrate: 128 kbps | Sample Rate: 44100 Hz | Duration: 1366.39 seconds | Channels: 2 | File Size: 21350.14KB |
                            
                        
                                
                                
                                
                                    
                                
                                
                                
                    
                
            Codec: mp3 | Bitrate: 128 kbps | Sample Rate: 44100 Hz | Duration: 1366.39 seconds | Channels: 2 | File Size: 21350.14KB |
                        Analytic Name: aasist
                        
                    
                                
                    
                            Description: Deep learning based method for Audio deepfake Detection, trained on ASV spoof dataset
                            
                    
                                Reference:
                                
                                    https://arxiv.org/abs/2110.01200
                                
                            
                            
                                Code Link:
                                
                                    https://github.com/clovaai/aasist
                                
                            
                        
                                
                                    🧪 Fake Probability: 97.3%
                                
                                
                                
                                
                                    ▲
                                    
                                
                                    97.3%
                                
                                
                                AI-generated Likelihood
                                
                        
                        
                        
                            Scope: This analytic can be used to detect variety deepfake audios, include voice conversion and TTS.
                        
                    