SavedAdd your journal

Journals / Muş Alparslan Üniversitesi Fen Bilimleri Dergisi / 2026 / Cilt: 14 Sayı: 1

Noise Reduction in Audio Signals using Kalman and RLS Filters

Ses İşaretlerinde Kalman ve RLS Filtreleri Kullanarak Gürültünün Azaltılması

Merve Şeyma Arslan(İstanbul su ve Kanalizasyon İdaresi Genel Müdürlüğü)0009-0008-5394-6322
Kübra Tancı(Tokat Gaziosmanpaşa Üniversitesi)0000-0002-1400-5043
Mahmut Hekim(TOKAT GAZİOSMANPAŞA ÜNİVERSİTESİ)0000-0002-6021-5703
Pages
38–46
Publication Date
2026-06-27
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

As a result, the experiments illustrated that the performance of both adaptive filters varied across different noise environments and SNR levels and that they were effective in reducing noise in speech signals.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This article focuses on comparing Kalman and recursive least squares (RLS) adaptive filter models used to reduce noise in speech signals.
📊
Key Findings & Data:To evaluate the performance of filter models, different experiments were repeated using noisy and noise-free datasets generated in both outdoor and indoor environments at different signal-to-noise ratio (SNR) levels.
💡
Conclusion & Impact:As a result, the experiments illustrated that the performance of both adaptive filters varied across different noise environments and SNR levels and that they were effective in reducing noise in speech signals.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Adaptif filtreleme · Gürültü azaltma · Kalman filtresi

Abstract

This article focuses on comparing Kalman and recursive least squares (RLS) adaptive filter models used to reduce noise in speech signals. For this aim, datasets containing noisy and noise-free reference female voice were recorded in both outdoor (traffic) and indoor (shopping mall) environments, and both filter models were designed for comparison using the Python programming language. To evaluate the performance of filter models, different experiments were repeated using noisy and noise-free datasets generated in both outdoor and indoor environments at different signal-to-noise ratio (SNR) levels. The performance experiments demonstrated that the filters were effective both in reducing noise and preserving signal integrity. The Kalman filter provided optimal estimation by modeling system dynamics, while the RLS filter stood out with its fast adaptation and low error rates. As a result, the experiments illustrated that the performance of both adaptive filters varied across different noise environments and SNR levels and that they were effective in reducing noise in speech signals.

Özet

Bu makalede, konuşma işaretlerindeki gürültüyü azaltmak için kullanılan Kalman ve yinelemeli en küçük kareler (RLS) adaptif filtre modellerinin karşılaştırılmasına odaklanılmıştır. Bu amaç için, dış ortamda (trafik) ve iç ortamda (alışveriş merkezi) gürültülü kadın sesi kayıtları ve gürültüsüz referans kayıtları içeren bir veri setleri oluşturulmuş ve karşılaştırma için Python programlama dili kullanılarak her iki filtre modeli tasarlanmıştır. Filtre modellerinin performanslarını değerlendirmek için, farklı işaret-gürültü oranı (SNR) seviyelerinde dış ve iç ortamda oluşturulan gürültülü ve gürültüsüz veri setleri kullanılarak deneyler tekrarlanmıştır. Gerçekleştirilen performans deneyleri, filtrelerin hem gürültünün azaltılmasında hem de işaret bütünlüğünün korunmasında etkin olduğunu göstermiştir. Kalman filtresi, sistem dinamiklerine uygun modelleme ile optimum tahmin sağlarken, RLS filtresi hızlı uyum kabiliyeti ve düşük hata oranlarıyla öne çıkmıştır. Sonuç olarak, gerçekleştirilen deneyler farklı gürültü ortamları ve SNR seviyelerinde her iki adaptif filtrenin de performanslarının değiştiğini ve konuşma işaretlerinde gürültünün azaltılması için etkili olduğunu göstermiştir.

🔖

Cite This Article

Arslan, M. Ş., Tancı, K., Hekim, M. (2026). Noise Reduction in Audio Signals using Kalman and RLS Filters. Muş Alparslan Üniversitesi Fen Bilimleri Dergisi, Cilt: 14 Sayı: 1, 38–46. https://doi.org/10.18586/msufbd.1858282

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (19)

  1. Haykin S. Communication Systems, 5th ed., Wiley, 2013.
  2. Somefun C.T., Daramola S.A., Somefun T.E. Advancements and applications of adaptive filters in signal processing, JESA, 57(05_02), 2024.
  3. Liu Y., Lei Z. Review of advances in active impulsive noise control with focus on adaptive algorithms, Applied Sciences, 14(3) 1218, 2024.
  4. Paliwal K.K., Basu A. A speech enhancement method based on Kalman filtering, Proceedings of ICASSP, 1 177–180, 1997.
  5. Özbek L. Gürültülü ses sinyali iyileştirilmesine ikili Kalman filtre yaklaşımı, Yüksek Lisans Tezi, Ankara Üniversitesi Fen Bilimleri Enstitüsü, 2007.
  6. Sayed A.H., Lotfizad M. Comparative analysis of adaptive noise cancellation techniques using LMS, NLMS and RLS algorithms, Proceedings of the International Conference on Signal Processing Systems, 140–144, 2011.
  7. Guimarães M.T., Ranzatto R.C.R., de Oliveira L.P. Speech enhancement based on Kalman filtering and wavelet transform, EURASIP Journal on Advances in Signal Processing, 2016(1) 1–10, 2016.
  8. Koning R., Bruce I.C., Denys S., Wouters J. Perceptual and model-based evaluation of ideal time-frequency noise reduction in hearing-impaired listeners, IEEE Transactions on Neural Systems and Rehabilitation Engineering, 26(3) 687–697, 2018.
  9. Marquardt D., Doclo S. Interaural coherence preservation for binaural noise reduction using partial noise estimation and spectral postfiltering, IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(7) 1261–1274, 2018.
  10. Andersen K.T., Moonen M. Robust speech-distortion weighted interframe Wiener filters for single-channel noise reduction, IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(1) 97–107, 2018.
  11. Dietzen T., Doclo S., Moonen M., van Waterschoot T. Integrated sidelobe cancellation and linear prediction Kalman filter for joint multi-microphone speech dereverberation, interfering speech cancellation, and noise reduction, IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28 740–754, 2020.
  12. Balaji V., Santhosh Kumar A., Sumathi P. Comparative study of adaptive filtering techniques for speech enhancement in noisy environments, International Journal of Engineering and Advanced Technology, 9(3) 2652–2656, 2020.
  13. Xue W., Zhang X., Wang D. Neural Kalman filtering for speech enhancement, IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28 2291–2302, 2020.
  14. Xue W., Moore A.H., Brookes M., Naylor P.A. Speech enhancement based on modulation-domain parametric multichannel Kalman filtering, IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29 393–405, 2021.
  15. Çolak R., Akdeniz R. A novel voice activity detection for multi-channel noise reduction, IEEE Access, 9 91017–91026, 2021.
  16. Sahael A., Djemal R., Ellouze N. An adaptive Kalman filtering approach for speech enhancement in real noisy environments, Journal of Electrical Engineering & Technology, 16(2) 805–814, 2021.
  17. Seidel M., Birkholz P., Uncini A. Deep learning-assisted frequency-domain Kalman filter for acoustic echo cancellation, IEEE Transactions on Audio, Speech, and Language Processing, 33(4) 778–789, 2025.
  18. Kalman R.E. A new approach to linear filtering and prediction problems, Journal of Basic Engineering, 82(D) 35–45, 1960.
  19. Lai, B., & Bernstein, D. S. (2024). Adaptive Kalman filtering developed from recursive least squares forgetting algorithms. In Proceedings of the American Control Conference (ACC) (pp. 4363–4368). Toronto, Canada.