widevine-l3-decryptor
fairseq
Our great sponsors
widevine-l3-decryptor | fairseq | |
---|---|---|
15 | 89 | |
878 | 29,205 | |
- | 1.6% | |
2.4 | 6.6 | |
over 2 years ago | 7 days ago | |
Python | ||
- | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
widevine-l3-decryptor
-
Ask HN: How does Google Widevine work under the hood?
It's a part of the browser. It's not doing it with Javascript if that's what you're asking. Chrome includes a file with the name widevinecdm.dll or something like that on Windows. No one knows exactly what this file does because it is incredibly obfuscated https://github.com/tomer8007/widevine-l3-decryptor/wiki/Reve....
As for what Widevine actually does, it just uses a protobuf based protocol to request a decryption key from a license server. License request messages from the client have to be signed with a valid device private key, which are made difficult to extract but some occasionally leak.
-
The Quest for Netflix on Asahi Linux
Not just privately either, there have been tools circulating even on GitHub.
For a L3 example there's one repo [1] that's kind of still up but not really. Still enough to show that it happened. L1 bypass has also been on GitHub briefly. However these things get deleted rather fast for obvious reasons.
--
[1] https://github.com/tomer8007/widevine-l3-decryptor
-
Decompiling x86 Deep Neural Network Executables
Widevine has been broken at least a few times, including by recovering the private key from its white-box implementation: https://github.com/tomer8007/widevine-l3-decryptor/wiki/Reve.... Note that the write-up says it was the "old" version, but that's relative to the date of the write-up. Google overhauled Widevine after he broke it.
I'm less familiar with shielding data like this, but historically things like VMProtect and Themida were the standard for shielding programs. These offer a degree of resistance to automation, but a determined human will eventually figure them out, and then automation usually follows anyway. Syntia did this for VMProtect and Themida: https://www.usenix.org/system/files/conference/usenixsecurit....
-
How to do webrips and webdl?
For webrip you'll need an hdmi splitter that doesnt abide by HDCP to bypass protections into a capture card. For webdl theres this older github but it doesn't work anymore, im not sure if there's a currently working script and key online publicly https://github.com/tomer8007/widevine-l3-decryptor
- Reversing the Old Widevine Content Decryption Module
-
Enabling scroll lock at boot fixes crashes caused by DRM on Intel 12th gen CPUs
There's code for the L3, but it doesn't work since end of may because the new RSA keys aren't public
https://github.com/tomer8007/widevine-l3-decryptor
Find the code in the forks, since it was DMCA'd
There's none for L1 and L2 though
-
Archiving Github Repos?
So if for example I `git clone` ed this project back when it was live with the original files, and then `git pull ` now after the maintainer removed project files and all the relevant commits with it, the `git pull` command would just fail?
-
reddit is forcing the SaveVideo bot offline or face legal action.
You can still bypass DRM if you know how or have to programms to do it (With widevine-l3-decryptor for example (The offical key(s) was/were revoked but some people still post keys))
fairseq
- Sequence-to-Sequence Toolkit Written in Python
- Unsupervised (Semi-Supervised) ASR/STT training recipes
-
Nvidia's 900 tons of GPU muscle bulks up server market, slims down wallets
> Is there really no way to partition the workload to run with 16gb memory per card?
It really depends and this can get really complicated really fast. I'll give a tldr and then a longer explanation.
TLDR:
Yes, you can easily split networks up. If your main bottleneck is batch size (i.e. training) then there aren't huge differences in spreading across multiple GPUs assuming you have good interconnects (GPU direct is supported). If you're running inference and the model fits on the card you're probably fine too unless you need to do things like fancy inference batching (i.e. you have LOTS of users)
Longer version:
You can always split things up. If we think about networks we recognize some nice properties about how they operate as mathematical groups. Non-residual networks are compositional, meaning each layer can be treated as a sub network (every residual block can be treated this way too). Additionally, we may have associative and distributive properties depending on the architecture (some even have commutative!). So we can use these same rules to break apart networks in many different ways. There are often performance hits for doing this though, as it practically requires you touching the disk more often but in some more rare cases (at least to me, let me know if you know more) they can help.
I mentioned the batching above and this can get kinda complicated. There are actually performance differences when you batch in groups of data (i.e. across GPUs) compared to batching on a single accelerator. This difference isn't talked about a lot. But it is going to come down to how often your algorithm depends on batching and what operations are used, such as batch norm. The batch norm is calculated across the GPU's batch, not the distributed batch (unless you introduce blocking). This is because your gradients AND inference are going to be computed differently. In DDP your whole network is cloned across cards so you basically run inference on multiple networks and then do an all reduce on the loss then calculate the gradient and then recopy the weights to all cards. There is even a bigger difference when you use lazy regularization (don't compute gradients for n-minibatches). GANs are notorious for using this and personally I've seen large benefits to distributed training for these. GANs usually have small batch sizes and aren't getting anywhere near the memory of the card anyways (GANs are typically unstable so large batch sizes can harm them), but also pay attention to this when evaluating papers (of course as well as how much hyper-parameter tuning has been done. This is always tricky when comparing works, especially between academia and big labs. You can easily be fooled by which is a better model. Evaluating models is way tougher than people give credit to and especially in the modern era of LLMs. I could rant a lot about just this alone). Basically in short, we can think of this as an ensembling method, except our models are actually identical (you could parallel reduce lazily too and that will create some periodic divergence between your models but that's not important for conceptually understanding, just worth noting).
There is are also techniques to split a single model up called model sharding and checkpointing. Model sharding is where you split a single model across multiple GPUs. You're taking advantage of the compositional property of networks, meaning that as long as there isn't a residual layer between your split location you can actually treat one network as a series of smaller networks. This has obvious drawbacks as you need to feed one into another and so the operations have to be synchronous, but sometimes this isn't too bad. Checkpointing is very similar but you're just doing the same thing on the same GPU. Your hit here is in I/O, but may or may not be too bad with GPU Direct and highly depends on your model size (were you splitting because batch size or because model size?).
This is all still pretty high level but if you want to dig into it more META developed a toolkit called fairseq that will do a lot of this for you and they optimized it
https://engineering.fb.com/2021/07/15/open-source/fsdp/
https://github.com/facebookresearch/fairseq
TLDR: really depends on your use case, but it is a good question.
-
Talk back and forth with AI like you would with a person
How do they do the text to voice conversion so fast? https://github.com/facebookresearch/fairseq/tree/main (open source takes sub-minute to do text to voice.
-
Voice generation AI (TTS)
It might be worth checking out Meta's TTS tho, I haven't gotten the chance to fiddle around with it but it looks somewhat promising https://github.com/facebookresearch/fairseq/tree/main/examples/mms
-
Translation app with TTS (text-to-speech) for Persian?
They have instructions on how to use it in command line and a notebook on how to use it as a python library.
- Why no work on open source TTS (Text to speech) models
-
Meta's Massively Multilingual Speech project supports 1k languages using self supervised learning
Github - https://github.com/facebookresearch/fairseq/tree/main/examples/mms Paper - https://research.facebook.com/publications/scaling-speech-technology-to-1000-languages/
-
AI — weekly megathread!
Meta released a new open-source model, Massively Multilingual Speech (MMS) that can do both speech-to-text and text-to-speech in 1,107 languages and can also recognize 4,000+ spoken languages. Existing speech recognition models only cover approximately 100 languages out of the 7,000+ known spoken languages. [Details | Research Paper | GitHub].
- Meta's MMS: Scaling Speech Technology to 1000+ languages (How to Run colab)
What are some alternatives?
yt-dlp - A feature-rich command-line audio/video downloader
gpt-neox - An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.
node - Node.js JavaScript runtime ✨🐢🚀✨
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
youtube-dl-server - Web / REST interface for downloading youtube videos onto a server.
DeepSpeed - DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
dumper - Dump L3 CDM from any Android device
text-to-text-transfer-transformer - Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"
reveddit - Review removed content on reddit. Uses the Pushshift API, built on code from removeddit.
espnet - End-to-End Speech Processing Toolkit
BunkerWeb - 🛡️ Make your web services secure by default !
Pytorch - Tensors and Dynamic neural networks in Python with strong GPU acceleration