paper-analysis-assistant

Paper Analysis Assistant

Task Objective

• This Skill is used to: automatically perform multidimensional analysis based on an arXiv paper URL and generate outputs in multiple formats
• Capabilities include: PDF download and text extraction, word frequency statistics, speech synthesis, podcast dialogue generation, interactive webpage, PPT generation, summary graphic generation, citation analysis
• Trigger conditions: the user provides an arXiv paper URL or a paper PDF file

Prerequisites

• Dependency description: the required Python packages are listed in dependency
• Stopword resource: an English stopword list needs to be prepared for filtering in word frequency analysis

Operation Steps

• Standard process:

  1. Download the PDF file
     • Call scripts/download_pdf.py to download the arXiv PDF
     • Parameters:--url (arXiv paper URL), --output (output PDF file path)
  2. Extract PDF text
     • Call scripts/extract_text.py to extract plain text
     • Parameters:--pdf (PDF file path), --output (output txt file path)
  3. Word frequency analysis
     • Call scripts/analyze_word_frequency.py to perform word frequency statistics
     • Parameters:--txt (txt file path), --output (output csv file path)
     • This script automatically filters English stopwords (see references/stopwords.txt)
  4. Text to speech
     • Call scripts/text_to_speech.py to convert text to speech
     • Parameters:--txt (txt file path), --output (output wav file path)
  5. Generate podcast dialogue
     • Agent step: generate a two-person dialogue script based on the paper content (including dialogue content for two roles)
     • Script step: call scripts/dialogue_to_podcast.py to convert the dialogue script to speech
     • Parameters:--dialogue (dialogue script file path), --output (output wav file path)
  6. Generate interactive webpage
     • Call scripts/generate_html.py to generate an interactive webpage
     • Parameters:--txt (txt file path), --word_freq (word frequency csv file path), --output (output html file path)
  7. Generate PPT
     • Call scripts/generate_ppt.py to generate a presentation
     • Parameters:--txt (txt file path), --output (output pptx file path)
  8. Generate summary graphic
     • Agent step: directly generate a "a picture is worth a thousand words" summary graphic based on the paper content (PNG format)
  9. Analyze citation links
     • Call scripts/extract_references.py Extract citation links
     • Parameters:--txt (txt file path), --output (output csv file path)

• Optional branches:

  • When the user directly provides a PDF file: skip step 1 and start directly from step 2
  • When the user only needs partial analysis: selectively execute the corresponding steps according to the requirements

Resource Index

• Download script: see scripts/download_pdf.py(Purpose: download arXiv PDF)
• Text extraction: see scripts/extract_text.py(Purpose: extract plain text from PDF)
• Word frequency analysis: see scripts/analyze_word_frequency.py(Purpose: count word frequencies and filter stop words)
• Speech synthesis: see scripts/text_to_speech.py(Purpose: text-to-speech)
• Podcast generation: see scripts/dialogue_to_podcast.py(Purpose: convert dialogue script to speech)
• Web page generation: see scripts/generate_html.py(Purpose: generate interactive HTML)
• PPT generation: see scripts/generate_ppt.py(Purpose: generate a PPT presentation)
• Citation extraction: see scripts/extract_references.py(Purpose: extract citation links)
• Stop word list: see references/stopwords.txt(Purpose: stop word filtering during word frequency analysis)

Notes

• Ensure that all script parameter paths are correct, especially the relative paths of input and output files
• The speech synthesis function requires system support for a speech engine (pyttsx3)
• The script content of the podcast dialogue is generated by the agent, and the dialogue format must be ensured to be correct
• The summary diagram is generated directly by the agent, with no need to call a script

Usage Examples

• Example 1: Complete analysis workflow

  # 下载 PDF
  python scripts/download_pdf.py --url "https://arxiv.org/abs/2301.00001" --output ./user-data/paper.pdf
  # 提取文本
  python scripts/extract_text.py --pdf ./user-data/paper.pdf --output ./user-data/paper.txt
  # 词频分析
  python scripts/analyze_word_frequency.py --txt ./user-data/paper.txt --output ./user-data/word_freq.csv
  # 语音合成
  python scripts/text_to_speech.py --txt ./user-data/paper.txt --output ./user-data/paper.wav
  # 播客对话（智能体生成对话脚本后）
  python scripts/dialogue_to_podcast.py --dialogue ./user-data/dialogue.txt --output ./user-data/podcast.wav
  # 生成网页
  python scripts/generate_html.py --txt ./user-data/paper.txt --word_freq ./user-data/word_freq.csv --output ./user-data/analysis.html
  # 生成 PPT
  python scripts/generate_ppt.py --txt ./user-data/paper.txt --output ./user-data/presentation.pptx
  # 提取引用
  python scripts/extract_references.py --txt ./user-data/paper.txt --output ./user-data/references.csv
  

• Example 2: Quick analysis (word frequency and citations only)

  python scripts/analyze_word_frequency.py --txt ./user-data/paper.txt --output ./user-data/word_freq.csv
  python scripts/extract_references.py --txt ./user-data/paper.txt --output ./user-data/references.csv