A collection of utilities for transcribing, summarizing, classifying, and creating visualizations from transcripts and summaries of various podcasts associated with the Intellectual Dark Web, conspiracy theories, QAnon, the Alt-Right, White Supremacist/Nationalist movements, and the Manosphere.
Download a given podcast's episodes and descriptions, or fetch transcripts from an API.
Examples:
python agloop.py --episodes "https://fight.fudgie.org/search/api/shows/sf/episodes" --transcripts vtt
python pill-feeder.py https://feeds.simplecast.com/vsy1m5LV
telechargeur configs/tucker.txt
Watch a given podcast's media directory to detect new episodes and transcribe them with Whisper.
Examples:
python careless-whisper-pill.py "/data/The Roseanne Barr Podcast/mp3"
python les-observateurs.py configs/roseanne.toml
Generate a variety of classifications based on transcripts or generated summaries.
Examples:
python EMOTIONAL-DAMAGE.py 1mjcwuaIJtW_9bGAebM3QK8RltWD9bKrjcr3qgMpivog
python zero-shot-thirty.py --candidate-labels "gender,feminism,politics,religion" "/data/The Tucker Carlson Show/vtt" tucker-zero-shot.csv
python entity-matrix.py "/data/Fresh & Fit/vtt" digfemnet.json 1ZTUTmzyko7hTLsiokXoV-eliUujmazElQ1bET_1234
Generate visualizations from the transcripts or summaries of one or more podcasts.
Examples:
python emotional-roller-coaster.py "Joe Rogan Experience #1509 - Abigail Shrier.vtt" "emotion-heatmap-joe-rogan-experience-1509-abigail-shrier.html" --title "Emotions: Joe Rogan Experience #1509 - Abigail Shrier"
python red-pill-caliper.py "/data/The Culture War - Tim Pool/vtt" --podcast-name "The Culture War - Tim Pool"
python red-pill-cloud.py "/data/The Culture War - Tim Pool/vtt"the-culture-war-podcast-with-tim-pool.png --width 2560 --height 1440 --title "The Culture War Podcast with Tim Pool" --additional-stopwords="think,know"
python red-pill-emotional-damage.py 1oIa4jk5DagHvqpclM3j5kkK2dagG3FL42VRL5n3jIdM 1HJYZPfOoChrlvv9TyeZib74au-RMDciHI6TKB9gTiks emotions-matt-gaetz-donald-trump-jr.html --title "Emotions of Triggered - Donald Trump Jr & Firebrand - Matt Gaetz (j-hartmann/emotion-english-distilroberta-base)"
python red-pill-resonator.py --keywords "democrat*,republic*,deep state" --width 2560 --height 1440 --title "Keyword Trend: 'democrat*,republic*,deep state'" keyword-trend-democrat-republican-deep-state.html
python wave-of-misogyny.py "/data/Fresh & Fit/vtt/NFL Player Speech Valid or Misogynistic?.vtt" "misogyny-wave-nfl-player-speech-valid-or-misogynistic.html" --title "Fresh & Fit: NFL Player Speech Valid or Misogynistic? (MilaNLProc/bert-base-uncased-ear-misogyny)"
python dicks-hate-the-police.py "/data/America First - Nicholas J. Fuentes/vtt/Crisis Actor Greta Thurnberg Rallies Around Climate Change HOAX | America First Ep. 465.vtt" "hatespeech-crisis-actor-greta-thurnberg-rallies-around-climate-change-hoax-america-first-ep-465.html" --title "America First: Crisis Actor Greta Thurnberg Rallies Around Climate Change HOAX | America First Ep. 465 (facebook/roberta-hate-speech-dynabench-r4-target)"
python donut-hate.py --title "America First - Nicholas J. Fuentes" "/data/America First - Nicholas J. Fuentes/vtt" hate-speech-america-first-nicholas-j-fuentes.html
python donut-hate-women.py --title "Tate Speech" "/data/Tate Speech/vtt" misogyny-tate-speech.html
Generate summarizations of a given podcast from a directory of transcript files.
Examples:
python recap-in-the-sheets.py "/data/The Joe Rogan Experience" 1mjcwuaIJtW_9bGAebM3QK8RltWD9bKrjcr3qgMpivog digfemnet.json
python redpill-recap.py "/data/The Joe Rogan Experience/vtt" "/data/The Joe Rogan Experience/summarizations"
python redpill-recap-stats.py "/data/The StoneZONE with Roger Stone/vtt/" roger-transcript-stats.csv
The Unlicense
This project is part of the Digital Feminist Network and is funded by the Social Sciences and Humanities Research Council. Additional financial and in-kind support comes from York University, York University Libraries, the Faculty of Arts, and the David R. Cheriton School of Computer Science at the University of Waterloo.