- This tool will not work without decryption keys. Do not bother installing unless you already have keys or can obtain them!
- If you ask about keys in the issues, your message will be deleted and you will be blocked.
- Downloading courses is against Udemy's Terms of Service, I am NOT held responsible for your account getting suspended as a result from the use of this program!
- This program is WIP, the code is provided as-is and I am not held resposible for any legal issues resulting from the use of this program.
Utility script to download Udemy courses, has support for DRM videos but requires the user to acquire the decryption key (for legal reasons).
Windows is the primary development OS, but I've made an effort to support Linux also (Mac untested).
Caution
The ability to download captions automatically is currently broken due to changes in Udemy's API!
Important
This tool will not work on encrypted courses without decryption keys being provided!
Downloading courses is against Udemy's Terms of Service, I am NOT held responsible for your account getting suspended as a result from the use of this program!
This program is WIP, the code is provided as-is and I am not held resposible for any legal issues resulting from the use of this program.
The following are a list of required third-party tools, you will need to ensure they are in your systems path and that typing their name in a terminal invokes them.
Note
These are seperate requirements that are not installed with the pip command!
You will need to download and install these manually!
- Python 3
- ffmpeg - This tool is also available in Linux package repositories.
- NOTE: It is recommended to use a custom build from the yt-dlp team that contains various patches for issues when used alongside yt-dlp, however it is not required. Latest builds can be found here
- aria2/aria2c - This tool is also available in Linux package repositories
- shaka-packager
- yt-dlp - This tool is also available in Linux package repositories, but can also be installed using pip if desired (
pip install yt-dlp)
You will need to get a few things before you can use this program:
- Decryption Key ID
- Decryption Key
- Udemy Course URL
- Udemy Bearer Token (aka acccess token for udemy-dl users)
- Udemy cookies (only required for subscription plans - see Udemy Subscription Plans)
- rename
.env.sampleto.env(you only need to do this if you plan to use the .env file to store your bearer token) - rename
keyfile.example.jsontokeyfile.json
- Firefox: Udemy-DL Guide
- Chrome: Udemy-DL Guide
- If you want to use the .env file to store your Bearer Token, edit the .env and add your token.
Important
For courses that are encrypted, It is up to you to acquire the decryption keys.
Please DO NOT ask me for help acquiring these!
- Enter the key and key id in the
keyfile.json
Tip
Cookies are not required for individually purchased courses.
To download a course included in a subscription plan that you did not purchase individually, you will need to use cookies. You can also use cookies as an alternative to Bearer Tokens.
The program can automatically extract them from your browser. You can specify what browser to extract cookies from with the --browser argument. Supported browsers are:
chromefirefoxoperaedgebravechromiumvivaldisafari
You can now run the program, see the examples below. The course will download to out_dir.
usage: main.py [-h] -c COURSE_URL [-b BEARER_TOKEN] [-q QUALITY] [-l LANG] [-cd CONCURRENT_DOWNLOADS] [--skip-lectures] [--download-assets]
[--download-captions] [--download-quizzes] [--keep-vtt] [--skip-hls] [--info] [--id-as-course-name] [-sc] [--save-to-file] [--load-from-file]
[--log-level LOG_LEVEL] [--browser {chrome,firefox,opera,edge,brave,chromium,vivaldi,safari}] [--use-h265] [--h265-crf H265_CRF] [--h265-preset H265_PRESET]
[--use-nvenc] [--out OUT] [--continue-lecture-numbers]
Udemy Downloader
options:
-h, --help show this help message and exit
-c COURSE_URL, --course-url COURSE_URL
The URL of the course to download
-b BEARER_TOKEN, --bearer BEARER_TOKEN
The Bearer token to use
-q QUALITY, --quality QUALITY
Download specific video quality. If the requested quality isn't available, the closest quality will be used. If not specified, the best quality will be
downloaded for each lecture
-l LANG, --lang LANG The language to download for captions, specify 'all' to download all captions (Default is 'en')
-cd CONCURRENT_DOWNLOADS, --concurrent-downloads CONCURRENT_DOWNLOADS
The number of maximum concurrent downloads for segments (HLS and DASH, must be a number 1-30)
--skip-lectures If specified, lectures won't be downloaded
--download-assets If specified, lecture assets will be downloaded
--download-captions If specified, captions will be downloaded
--download-quizzes If specified, quizzes will be downloaded
--keep-vtt If specified, .vtt files won't be removed
--skip-hls If specified, hls streams will be skipped (faster fetching) (hls streams usually contain 1080p quality for non-drm lectures)
--info If specified, only course information will be printed, nothing will be downloaded
--id-as-course-name If specified, the course id will be used in place of the course name for the output directory. This is a 'hack' to reduce the path length
-sc, --subscription-course
Mark the course as a subscription based course, use this if you are having problems with the program auto detecting it
--save-to-file If specified, course content will be saved to a file that can be loaded later with --load-from-file, this can reduce processing time (Note that asset
links expire after a certain amount of time)
--load-from-file If specified, course content will be loaded from a previously saved file with --save-to-file, this can reduce processing time (Note that asset links
expire after a certain amount of time)
--log-level LOG_LEVEL
Logging level: one of DEBUG, INFO, ERROR, WARNING, CRITICAL (Default is INFO)
--browser {chrome,firefox,opera,edge,brave,chromium,vivaldi,safari}
The browser to extract cookies from
--use-h265 If specified, videos will be encoded with the H.265 codec
--h265-crf H265_CRF Set a custom CRF value for H.265 encoding. FFMPEG default is 28
--h265-preset H265_PRESET
Set a custom preset value for H.265 encoding. FFMPEG default is medium
--use-nvenc Whether to use the NVIDIA hardware transcoding for H.265. Only works if you have a supported NVIDIA GPU and ffmpeg with nvenc support
--out OUT, -o OUT Set the path to the output directory
--continue-lecture-numbers, -n
Use continuous lecture numbering instead of per-chapter
- Passing a Bearer Token and Course ID as an argument
python main.py -c <Course URL> -b <Bearer Token>python main.py -c https://www.udemy.com/courses/myawesomecourse -b <Bearer Token>
- Download a specific quality
python main.py -c <Course URL> -q 720
- Download assets along with lectures
python main.py -c <Course URL> --download-assets
- Download assets and specify a quality
python main.py -c <Course URL> -q 360 --download-assets
- Download captions (Defaults to English)
python main.py -c <Course URL> --download-captions
- Download captions with specific language
python main.py -c <Course URL> --download-captions -l en- English subtitlespython main.py -c <Course URL> --download-captions -l es- Spanish subtitlespython main.py -c <Course URL> --download-captions -l it- Italian subtitlespython main.py -c <Course URL> --download-captions -l pl- Polish Subtitlespython main.py -c <Course URL> --download-captions -l all- Downloads all subtitles- etc
- Skip downloading lecture videos
python main.py -c <Course URL> --skip-lectures --download-captions- Downloads only captionspython main.py -c <Course URL> --skip-lectures --download-assets- Downloads only assets
- Keep .VTT caption files:
python main.py -c <Course URL> --download-captions --keep-vtt
- Skip parsing HLS Streams (HLS streams usually contain 1080p quality for Non-DRM lectures):
python main.py -c <Course URL> --skip-hls
- Print course information only:
python main.py -c <Course URL> --info
- Specify max number of concurrent downloads:
python main.py -c <Course URL> --concurrent-downloads 20python main.py -c <Course URL> -cd 20
- Cache course information:
python main.py -c <Course URL> --save-to-file
- Load course cache:
python main.py -c <Course URL> --load-from-file
- Change logging level:
python main.py -c <Course URL> --log-level DEBUGpython main.py -c <Course URL> --log-level WARNINGpython main.py -c <Course URL> --log-level INFOpython main.py -c <Course URL> --log-level CRITICAL
- Use course ID as the course name:
python main.py -c <Course URL> --id-as-course-name
- Encode in H.265:
python main.py -c <Course URL> --use-h265
- Encode in H.265 with custom CRF:
python main.py -c <Course URL> --use-h265 -h265-crf 20
- Encode in H.265 with custom preset:
python main.py -c <Course URL> --use-h265 --h265-preset faster
- Encode in H.265 using NVIDIA hardware transcoding:
python main.py -c <Course URL> --use-h265 --use-nvenc
- Use continuous numbering (don't restart at 1 in every chapter):
python main.py -c <Course URL> --continue-lecture-numberspython main.py -c <Course URL> -n
if you want help using the program, join my Discord server or use GitHub Issues
- https://github.com/Jayapraveen/Drm-Dash-stream-downloader - For the original code which this is based on
- https://github.com/alastairmccormack/pywvpssh - For code related to PSSH extraction
- https://github.com/alastairmccormack/pymp4parse - For code related to mp4 box parsing (used by pywvpssh)
- https://github.com/lbrayner/vtt-to-srt - For code related to converting subtitles from vtt to srt format
- https://github.com/r0oth3x49/udemy-dl - For some of the informaton related to using the udemy api
All code is licensed under the MIT license
Woo, you made it this far!
I spend a lot of time coding things, and almost all of them are for nothing in return. When theres a lot of use of a program I make, I try to keep it updated, fix bugs, and even implement new features! But after a while, I do run out of motivation to keep doing it. If you like my work, and can help me out even a little, it would really help me out. If you are interested, you can find all the available options here. Even if you don't, thank you anyways!
