Detect fake FLAC files — find MP3/AAC/Opus transcodes disguised as lossless audio. Fast Rust CLI + web UI.
-
Updated
Sep 7, 2026 - Rust
Detect fake FLAC files — find MP3/AAC/Opus transcodes disguised as lossless audio. Fast Rust CLI + web UI.
Ensemble speaker verification achieving 97% accuracy - Intelligent fusion of MFCC+DTW (92%) and Resemblyzer CNN (94%) for voice authentication
Local-first steganography and DFIR workbench for guided file analysis, encrypted evidence handling, payload hiding/recovery, and reproducible research.
Audio analysis toolkit
Desktop GUI tool for detecting audio deepfakes using hand-crafted features + XGBoost/LightGBM ensemble. Segment-level analysis, heatmaps, PDF reports. Zero deep learning required.
A robust AI-powered web app to detect deepfake audio using Wav2Vec2 and FastAPI. Deployed on Hugging Face Spaces for the CyberSecurity Hackathon.
Free, open-source forensic audio workstation, non-destructive editing, spectral analysis, and a tamper-evident audit trail, the kind of tooling usually locked behind expensive proprietary software.
A web app that detects deepfake voices using an LLM forensic analyzer.
Cross-platform audio quality and integrity analysis with lightweight Quick Inspect, deep evidence, and the Oracle Engine.
Free, local-first and explainable AI-origin music analysis for artists, labels, platforms, rights teams and researchers.
Detecta manipulaciones en grabaciones de audio con IA: localiza el empalme y genera un informe pericial con hash SHA-256
Local-first audio forensic analyzer for spectral cutoffs, transcodes, clipping and discontinuities.
A FLAC steganography toolkit for encrypted payloads, audio forensics, and PCM embedding benchmarks.
Forensic toolkit for detecting AI-synthetic voices through Nonlinear Phenomena (NLP) and phase-space reconstruction.
A web-based Audio Forensic Intelligence Platform for audio authenticity verification, fingerprint generation, tamper detection, watermark validation, and forensic report generation using Django and Librosa.
Explainable deepfake voice detection — RawNet2 + Whisper + real-time browser inference. 95.2% accuracy, 4.19% EER on ASVspoof 2019.
Full-stack forensic speech analysis platform for detecting AI-generated vs human audio with real-time inference and explainable ML.
A Decoupled Tri-Modal Architecture for Real-Time Deepfake Interception via WebRTC and Late-Fusion.
AcousticSpace — an end-to-end deepfake audio & acoustic forensics platform that analyzes 118 acoustic features, RIR/reverberation patterns, background consistency, and voice characteristics using Python, Librosa, ML, FastAPI, and React.
To associate your repository with the audio-forensics topic, visit your repo's landing page and select "manage topics."