Skip to content

isinmelih/aec-pbfdaf-baseline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 

Repository files navigation

#🔊 AEC PBFDAF Baseline

This repository provides a real-time acoustic echo cancellation (AEC) baseline based on a Partitioned Block Frequency-Domain Adaptive Filter (PBFDAF).

The implementation is designed for research, benchmarking, and educational purposes, emphasizing measurable behavior and real-time constraints.


##🔑 Key Features

  • Real-time PBFDAF-based adaptive filtering
  • Double-talk detection
  • Online ERLE measurement and convergence statistics
  • Clear separation between core processing and GUI control

This project prioritizes clarity and correctness over product-level optimizations.


⚖️ Citation & Usage Etiquette

This repository is provided as a baseline and research reference for acoustic echo cancellation using PBFDAF.

If you use this codebase, architecture, or ideas in:

  • Academic publications
  • Theses
  • Technical reports
  • Blog posts
  • Derivative implementations

Please cite or clearly acknowledge this repository.
This repository is intended to be referenced as a baseline, not rebranded as an original end-to-end AEC system.

###✔️ Suggested Citation

isinmelih. AEC PBFDAF Baseline: Research-oriented implementation of partitioned-block frequency-domain adaptive filtering for acoustic echo cancellation. GitHub repository, 2026.


💬 Join the Discussion

Share feedback & improvements on metrics & performance plots → Click here


⬇️ Download Latest Release

The latest precompiled binaries for aec-pbfdaf-baseline are available on GitHub Releases.

Download and run the ZIP package:
Latest Release v0.1

The ZIP contains:

  • wasapi_aec.exe (core engine)
  • control_panel.exe (GUI launcher)
  • README.md with usage instructions
  • Example metrics / performance plots (if included)

##📝 Usage Notes

  • This project is not intended to claim novelty or state-of-the-art performance
  • Designed for learning, evaluation, and controlled experimentation
  • Modifications and extensions should be clearly documented by downstream users
  • While the Apache 2.0 license permits reuse, ethical academic practice requires transparent attribution