Scholarlink

In the dynamic landscape of digital content creation, particularly within the influencer community, the need for advanced monitoring tools has never been more critical. This article explores cutting-edge technologies such as speech recognition and object detection, powered by AI models like Whisper.AI and YOLOv8, to scrutinise influencer content. Our investigation reveals these tools expanding capabilities in extracting relevant data and metadata to identify potential compliance issues and content violations. Whisper.AI’s transcription accuracy, tested on English-language YouTube videos, showed near-perfect results, surpassing YouTube’s native service by including punctuation and enhancing readability. YOLOv8 models demonstrated distinct differences in operational efficiency and detection capabilities, with yolov8n providing rapid detection and yolov8x excelling in accuracy. This study highlights the potential of these technologies for content monitoring and suggests future integration of sentiment analysis for comprehensive content analysis.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Unveiling the Secrets: In-Depth Analysis of YouTube Video Data and Metadata Extraction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • João Bragança,
  • Fábio Silva

摘要

In the dynamic landscape of digital content creation, particularly within the influencer community, the need for advanced monitoring tools has never been more critical. This article explores cutting-edge technologies such as speech recognition and object detection, powered by AI models like Whisper.AI and YOLOv8, to scrutinise influencer content. Our investigation reveals these tools expanding capabilities in extracting relevant data and metadata to identify potential compliance issues and content violations. Whisper.AI’s transcription accuracy, tested on English-language YouTube videos, showed near-perfect results, surpassing YouTube’s native service by including punctuation and enhancing readability. YOLOv8 models demonstrated distinct differences in operational efficiency and detection capabilities, with yolov8n providing rapid detection and yolov8x excelling in accuracy. This study highlights the potential of these technologies for content monitoring and suggests future integration of sentiment analysis for comprehensive content analysis.