Melola Discovery

← Melola Discovery

Jurnal SINTA 1

Performance of speech enhancement models in video conferences: DeepFilterNet3 and RNNoise

Sinergi · Vol. 29 No. 2 (2025) · hlm. 269–278

DOI
10.22441/sinergi.2025.2.001
ISSN
2460-1217, 1410-2331
Terbit
2025-05-01
Penerbit
Universitas Mercu Buana
Bahasa
eng

Abstrak

As remote work and online education continue to gain prominence, the importance of clear audio communication becomes crucial. Deep Learning-based Speech Enhancement has emerged as a promising solution for processing data in noisy environments. In this study, we conducted an in-depth analysis of two speech enhancement models, RNNoise and DeepFilterNet3, selected for their respective strengths. DeepFilterNet3 leverages time-frequency masking with a Complex Mask filter, while RNNoise employs Recurrent Neural Networks with lower complexity. The performance evaluation in training revealed that RNNoise demonstrated impressive denoising capabilities, achieving low loss values, while DeepFilterNet3 showed superior generalization. Specifically, "DeepFilterNet3 (Pre-Trained)" exhibited the best overall performance, excelling in intelligibility and speech quality. RNNoise also performed well in subjective quality measures. Furthermore, we assessed the real-time processing efficiency of both models. Both RNNoise variants processed speech signals almost in real-time, whereas DeepFilterNet3, though slightly slower, remained efficient. The findings demonstrate significant improvements in speech quality, with "DeepFilterNet3 (Pre-Trained)" emerging as the top-performing model. The implications of this study have the potential to enhance video conference experiences and contribute to the improvement of remote work and online education.

Kata kunci

Jaringan kolaborasi para penulis

26 penulis · 63 tautan kepenulisan bersama · 36 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Mercu Buana University
  • Beijing Institute of Technology
  • Jakarta State Polytechnic
  • Universitas Esa Unggul
  • Kigali Independent University
  • National Defence University of Malaysia
  • Bandung Institute of Technology
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Trie Maya KadarinaMercu Buana University3
Arifiansyah ZodyUniversitas Esa Unggul2
Mirna YunitaBeijing Institute of Technology2
Eko IhsantoMercu Buana University1
Fina SupeginaMercu Buana University1
Muhammad Imam MuthahharMercu Buana University1
Asif KhanBeijing Institute of Technology1
Andi AdriansyahMercu Buana University1
Sabin HitimanaKigali Independent University1
Bugi Nur RohmanMercu Buana University1
Rahmat MuwardiMercu Buana University1
Asif KhanBeijing Institute of Technology1
Rachmat MuwardiMercu Buana University1
Khoirul AzizMercu Buana University1
Inggis Kurnia TrisiawanMercu Buana University1
Kadek Dwi SuyasminiMercu Buana University1
Rinto PriambodoMercu Buana University1
Riandini RiandiniJakarta State Polytechnic1
Sulis SetiowatiJakarta State Polytechnic1
Rika Novita WardhaniJakarta State Polytechnic1
Mohd Taufik JusohNational Defence University of Malaysia1
Ketty Siti SalamahMercu Buana University1
Arrival Dwi SentosaBandung Institute of Technology1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://publikasi.mercubuana.ac.id/index.php/sinergi/oai, pertama di-harvest 10 Nov 2026, terakhir terlihat 16 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Performance of speech enhancement models in video conferences: DeepFilterNet3 and RNNoise</dc:title>
	<dc:creator>Maulana, Muhammad Iqbal</dc:creator>
	<dc:creator>Raisul Akbar, Muhammad Fadhlillah</dc:creator>
	<dc:creator>Iklima, Zendi</dc:creator>
	<dc:subject xml:lang="en-US">Accuracy Metrics</dc:subject>
	<dc:subject xml:lang="en-US">DeepFilterNet3</dc:subject>
	<dc:subject xml:lang="en-US">RNNoise</dc:subject>
	<dc:subject xml:lang="en-US">Speech Enhancement</dc:subject>
	<dc:subject xml:lang="en-US">Speed Metrics</dc:subject>
	<dc:description xml:lang="en-US">As remote work and online education continue to gain prominence, the importance of clear audio communication becomes crucial. Deep Learning-based Speech Enhancement has emerged as a promising solution for processing data in noisy environments. In this study, we conducted an in-depth analysis of two speech enhancement models, RNNoise and DeepFilterNet3, selected for their respective strengths. DeepFilterNet3 leverages time-frequency masking with a Complex Mask filter, while RNNoise employs Recurrent Neural Networks with lower complexity. The performance evaluation in training revealed that RNNoise demonstrated impressive denoising capabilities, achieving low loss values, while DeepFilterNet3 showed superior generalization. Specifically, "DeepFilterNet3 (Pre-Trained)" exhibited the best overall performance, excelling in intelligibility and speech quality. RNNoise also performed well in subjective quality measures. Furthermore, we assessed the real-time processing efficiency of both models. Both RNNoise variants processed speech signals almost in real-time, whereas DeepFilterNet3, though slightly slower, remained efficient. The findings demonstrate significant improvements in speech quality, with "DeepFilterNet3 (Pre-Trained)" emerging as the top-performing model. The implications of this study have the potential to enhance video conference experiences and contribute to the improvement of remote work and online education.</dc:description>
	<dc:publisher xml:lang="en-US">Universitas Mercu Buana</dc:publisher>
	<dc:date>2025-05-01</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:format>application/msword</dc:format>
	<dc:format>application/vnd.openxmlformats-officedocument.wordprocessingml.document</dc:format>
	<dc:identifier>https://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/21917</dc:identifier>
	<dc:identifier>10.22441/sinergi.2025.2.001</dc:identifier>
	<dc:source xml:lang="en-US">SINERGI; Vol. 29 No. 2 (2025); 269-278</dc:source>
	<dc:source>2460-1217</dc:source>
	<dc:source>1410-2331</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/21917/9619</dc:relation>
	<dc:relation>https://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/21917/11678</dc:relation>
	<dc:relation>https://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/21917/11680</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2025 SINERGI</dc:rights>
</oai_dc:dc>