lecture-downloader 1.0.0

A comprehensive toolkit for downloading, merging, and transcribing lecture videos

Lecture Downloader

A Python toolkit for downloading, merging, transcribing, and embedding subtitles from lecture videos hosted on platforms like Canvas and Brightspace.

Use at your own risk. This tool is designed for educational purposes and should not be used to violate any terms of service or copyright laws.

Quick Start

30-Second Setup

# Install FFmpeg (required for video processing)
brew install ffmpeg  # macOS
# sudo apt install ffmpeg  # Ubuntu/Debian
# Windows: https://www.wikihow.com/Install-FFmpeg-on-Windows

# Install lecture downloader
pip install lecture-downloader

Obtaining Video URLs from Canvas/Brightspace

Implementation based off this reddit post

Using Video DownloadHelper Extension

1. Install Extension: Download VideoDownloadHelper
2. Navigate to Video: Go to your lecture video in Canvas/Brightspace
3. Start Playback: Click play to begin streaming
4. Extract URL: Click the extension icon (should be colored, not grey)
5. Copy URL: Click the three dots → "Copy URL"

For example, visiit Public Lecture sample, click play on a video, and copy the URL from the extension. To bulk download, paste it into a text file named lecture_links.txt, one URL per line.

Basic Usage

One-Command Pipeline

# Complete pipeline: download → merge → transcribe

pipeline_results = ld.process_pipeline(
    links="lecture_links.txt",  # Can also use: single URL string, ["url1", "url2"]
    titles="lecture_titles.json",  # Can also use: ["Title 1", "Title 2"], {"Module 1": ["Lecture 1"]}
    output_dir="lecture_processing",
    inject_subtitles=True,          # False to skip subtitle injection
    transcription_method="whisper", # "auto", "gcloud", "whisper"
    language="en",                  
)

Step-by-Step Commands

import lecture_downloader as ld

# Complete workflow in 3 commands
base_dir = "Lecture-Downloads"

# 1. Download lectures
results = ld.download_lectures(
    links="lecture_links.txt",  # Can also use: single URL string, ["url1", "url2"]
    titles="lecture_titles.json",  # Can also use: ["Title 1", "Title 2"], {"Module 1": ["Lecture 1"]}
    base_dir=base_dir,  # Creates Lecture-Downloads/lecture-downloads/
)

# 2. Merge videos by module with chapters
merged = ld.merge_videos(
    base_dir=base_dir,  # Auto-detects input from lecture-downloads/
)

# 3. Transcribe with Whisper (local, no setup required)
transcripts = ld.transcribe_videos(
    base_dir=base_dir,  # Auto-detects input from merged-lectures/
    method="whisper",  # "auto" detects best available method
    language="en",  # Language code for Whisper
    inject_subtitles=True,  # False to skip subtitle injection
)

Installation

# Basic installation
pip install lecture-downloader

Required Dependencies:

• ffmpeg - Install via package manager (brew, apt, etc.)
• Python 3.8+

Configuration Options

Download Parameters

Parameter	Type	Default	Description
links	str/list	Required	File path, single URL, or list of URLs
titles	str/list/dict	None	File path, list, or dict mapping
base_dir	str	"."	Base directory (creates subdirectories)
max_workers	int	5	Concurrent downloads (1-10)
verbose	bool	False	Detailed progress output

Merge Parameters

Parameter	Type	Default	Description
base_dir	str	"."	Base directory (auto-detects input)
verbose	bool	False	Detailed progress output

Transcribe Parameters

Parameter	Type	Default	Description
base_dir	str	"."	Base directory (auto-detects input)
method	str	"auto"	"auto", "whisper"
language	str	"en"	Language code for Whisper
max_workers	int	3	Concurrent transcriptions (1-5)
inject_subtitles	bool	True	Inject SRT into video files
verbose	bool	False	Detailed progress output

Input Formats

Links Input

# File with URLs (one per line)
links = "lecture_links.txt"

# Single URL
links = "https://example.com/lecture.mp4"

# List of URLs
links = ["https://url1.mp4", "https://url2.mp4"]

Titles Input

# JSON file with module structure
titles = "lecture_titles.json"
**lecture_links.txt:**

https://example.com/lecture1.mp4 https://example.com/lecture2.mp4

# List of titles (matches link order)
titles = ["Lecture 1", "Lecture 2", "Lecture 3"]

# Dictionary mapping modules to lectures
titles = {
    "Module 1: Introduction": ["Lecture 1", "Lecture 2"],
    "Module 2: Advanced": ["Lecture 3", "Lecture 4"]}

lecture_titles.json:

{ "Module 1: Introduction": [ "Lecture 1: Overview", "Lecture 2: Fundamentals"], 
  "Module 2: Advanced Topics": [  "Lecture 3: Advanced Concepts"]}

CLI Usage

Quick Commands

# Complete workflow
BASE_DIR="Lecture-Downloads"
lecture-downloader download -l links.txt -t titles.json -b $BASE_DIR
lecture-downloader merge -b $BASE_DIR
lecture-downloader transcribe -b $BASE_DIR

# One-command pipeline
lecture-downloader pipeline -l links.txt -t titles.json -o output

CLI Options

# Download with options
lecture-downloader download \
  -l links.txt \
  -t titles.json \
  -b Lecture-Downloads \
  --max-workers 8 \
  --verbose

# Transcribe with options
lecture-downloader transcribe \
  -b Lecture-Downloads \
  --method whisper \
  --language en \
  --max-workers 4 \
  --no-inject

FFmpeg not found:

# Install FFmpeg
brew install ffmpeg  # macOS
sudo apt install ffmpeg  # Ubuntu/Debian

Debug Mode

# Enable verbose output for troubleshooting
ld.download_lectures(links, titles, verbose=True)
ld.merge_videos(base_dir="course", verbose=True)
ld.transcribe_videos(base_dir="course", verbose=True)

License

MIT License - see LICENSE file for details.