EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Automatic Detection of Toxic Content in Short Videos Using Deep Learning-Based Text and Audio Feature Integration

Heri Agus Supriyanto, Ryan Ari Setyawan, Jemmy Edwin Bororing

Journal of Deep Learning, Computer Vision and Digital Image Processing · 08.07.2026 · 221-235

DOI: 10.61255/decoding.v4i2.1540 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Deep Learning, Computer Vision and Digital Image Processing, online erschienen am 08.07.2026. Verfasst von Heri Agus Supriyanto und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose – This study develops and evaluates a multimodal classification model combining audio and text features via configurable weighted late fusion to detect toxic content in keyword-retrieved Indonesian short videos containing profanity-related contexts, addressing text-only detection limitations.Methods – The pipeline utilized FFmpeg for audio extraction, MFCC for audio features, and Google Speech Recognition for text. Three fusion configurations (Audio-Text: 40%–60%, 50%–50%, and 60%–40%) combining a DNN and BiLSTM were evaluated on 1,484 manually labeled Indonesian short videos.Findings – The Audio 60%–Text 40% configuration achieved the numerically highest test accuracy of 93.94% with a 95% confidence interval of 90.62%–96.13%, using a decision threshold of 0.60 selected from the validation set. The model obtained an F1-score of 0.95 for the toxic class. Compared with unimodal baselines, all fusion models achieved higher accuracy, indicating the benefit of integrating audio and text features.Research implications – The findings suggest that multimodal audio-text fusion can improve Indonesian short-form video toxicity detection compared with audio-only or text-only models. However, the differences among the three fusion weighting schemes were not statistically significant based on McNemar’s test, so the audio-dominant configuration should be interpreted as the numerically best configuration in this dataset rather than as a universally superior setting.Originality – This study systematically compares unimodal baselines and configurable audio-text late-fusion weighting strategies for Indonesian short-form video toxicity detection. The study also applies validation-based threshold selection, confidence intervals, and pairwise McNemar testing to provide a more reliable evaluation of multimodal model performance.

Zitieren

APA

Supriyanto, H. A., Setyawan, R. A., & Bororing, J. E. (2026). Automatic Detection of Toxic Content in Short Videos Using Deep Learning-Based Text and Audio Feature Integration. Journal of Deep Learning, Computer Vision and Digital Image Processing, 221-235. https://doi.org/10.61255/decoding.v4i2.1540

Vancouver

Supriyanto HA, Setyawan RA, Bororing JE. Automatic Detection of Toxic Content in Short Videos Using Deep Learning-Based Text and Audio Feature Integration. Journal of Deep Learning, Computer Vision and Digital Image Processing. 2026:221-235. doi:10.61255/decoding.v4i2.1540

BibTeX
@article{Supriyanto2026,
  author = {Supriyanto, Heri Agus and Setyawan, Ryan Ari and Bororing, Jemmy Edwin},
  title = {Automatic Detection of Toxic Content in Short Videos Using Deep Learning-Based Text and Audio Feature Integration},
  journal = {Journal of Deep Learning, Computer Vision and Digital Image Processing},
  year = {2026},
  pages = {221--235},
  doi = {10.61255/decoding.v4i2.1540}
}
Kostenlos starten →