Ensemble speaker verification achieving 97% accuracy - Intelligent fusion of MFCC+DTW (92%) and Resemblyzer CNN (94%) for voice authentication
deep-learning identity-verification ensemble-learning fraud-prevention speaker-recognition speaker-verification synthetic-speech-detection liveness-detection speaker-identification speaker-embedding voice-authentication voice-biometrics anti-spoofing audio-forensics biometric-security authentication-systems deepfake-audio-detection voice-forensics voice-security passive-authentication
-
Updated
Nov 13, 2025 - Jupyter Notebook