Ensemble speaker verification achieving 97% accuracy - Intelligent fusion of MFCC+DTW (92%) and Resemblyzer CNN (94%) for voice authentication
-
Updated
Nov 13, 2025 - Jupyter Notebook
Ensemble speaker verification achieving 97% accuracy - Intelligent fusion of MFCC+DTW (92%) and Resemblyzer CNN (94%) for voice authentication
Momenta's Speech DF Arena benchmark results: per-utterance score files for all 14 datasets, Arena protocol files, and a script to reproduce the 0.89% pooled EER / 1.01% average EER.
The missing trust layer for the open business communications stack. Antifraud toolkit for EU SMEs · EUPL-1.2 · NLnet NGI Zero Commons proposal.
Pindrop is a voice and audio security company providing a continuous identity verification and deepfake-detection platform for enterprises. Its technology authenticates callers and video-meeting participants, detects synthetic (AI-generated) speech and fraudulent activity, and protects contact centers, virtual meetings, and connected devices.
Hiya is an AI-native voice security company that blocks spam, fraud, and deepfake calls at network scale for mobile operators, enterprises, and consumers. Its developer platform exposes several REST APIs: Hiya Connect for branded outbound calling (registering originating numbers, managing branded phone numbers, logos, and managed brands), Hiya…
Add a description, image, and links to the voice-security topic page so that developers can more easily learn about it.
To associate your repository with the voice-security topic, visit your repo's landing page and select "manage topics."