xla
pocketsphinx
xla | pocketsphinx | |
---|---|---|
8 | 6 | |
2,296 | 3,745 | |
1.4% | 0.9% | |
9.9 | 7.4 | |
1 day ago | about 2 months ago | |
C++ | C | |
GNU General Public License v3.0 or later | GNU General Public License v3.0 or later |
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.
xla
-
Who uses Google TPUs for inference in production?
> The PyTorch/XLA Team at Google
Meanwhile you have an issue from 5 years ago with 0 support
https://github.com/pytorch/xla/issues/202
-
Google TPU v5p beats Nvidia H100
PyTorch has had an XLA backend for years. I don't know how performant it is though. https://pytorch.org/xla
-
Why Did Google Brain Exist?
It's curtains for XLA, to be precise. And PyTorch officially supports XLA backend nowadays too ([1]), which kind of makes JAX and PyTorch standing on the same foundation.
1. https://github.com/pytorch/xla
-
Accelerating AI inference?
Pytorch supports other kinds of accelerators (e.g. FPGA, and https://github.com/pytorch/glow), but unless you want to become a ML systems engineer and have money and time to throw away, or a business case to fund it, it is not worth it. In general, both pytorch and tensorflow have hardware abstractions that will compile down to device code. (XLA, https://github.com/pytorch/xla, https://github.com/pytorch/glow). TPUs and GPUs have very different strengths; so getting top performance requires a lot of manual optimizations. Considering the the cost of training LLM, it is time well spent.
-
[D] Colab TPU low performance
While apparently TPUs can theoretically achieve great speedups, getting to the point where they beat a single GPU requires a lot of fiddling around and debugging. A specific setup is required to make it work properly. E.g., here it says that to exploit TPUs you might need a better CPU to keep the TPU busy, than the one in colab. The tutorials I looked at oversimplified the whole matter, the same goes for pytorch-lightning which implies switching to TPU is as easy as changing a single parameter. Furthermore, none of the tutorials I saw (even after specifically searching for that) went into detail about why and how to set up a GCS bucket for data loading.
- How to train large deep learning models as a startup
-
Distributed Training Made Easy with PyTorch-Ignite
XLA on TPUs via pytorch/xla.
-
[P] PyTorch for TensorFlow Users - A Minimal Diff
I don't know of any such trick except for using TensorFlow. In fact, I benchmarked PyTorch XLA vs TensorFlow and found that the former's performance was quite abysmal: PyTorch XLA is very slow on Google Colab. The developers' explanation, as I understood it, was that TF was using features not available to the PyTorch XLA developers and that they therefore could not compete on performance. The situation may be different today, I don't know really.
pocketsphinx
- [Discussion] Looking for an Open-Source Speech to Text model (english) that captures filler words, pauses and also records timestamps for each word.
-
I Created A Web Speech API NPM Package Called SpeechKit
There are espeak-ng https://github.com/espeak-ng/espeak-ng and pocketsphinx https://github.com/cmusphinx/pocketsphinx which can be used locally without making external requests.
-
"Why not just transcribe the audio?" I thought
And so I installed PocketSphinx, "one of Carnegie Mellon University's open source large vocabulary, speaker-independent continuous speech recognition engines."
-
How to train large deep learning models as a startup
- https://github.com/cmusphinx/pocketsphinx
This avoids having to stream audio 24x7 to a cloud model which would be super expensive. This being said, I'm pretty sure what the Alexa does, for example, is send any positive wake word to a cloud model (that is bigger and more accurate) to verify the prediction of the local wake word detection model AFAIK.
- Speech recognition library for financial markets
-
Speech recognition
PocketSphinx is generally regarded among voice assistant communities as a less reliable, but straight OOTB, alternative to a robust listener. It's a good solution when you want multiple hotwords (or just aren't in a position to train even one word.)
What are some alternatives?
NCCL - Optimized primitives for collective multi-GPU communication
vosk - VOSK Speech Recognition Toolkit
pytorch-lightning - Build high-performance AI models with PyTorch Lightning (organized PyTorch). Deploy models with Lightning Apps (organized Python to build end-to-end ML systems). [Moved to: https://github.com/Lightning-AI/lightning]
snowboy - Future versions with model training module will be maintained through a forked version here: https://github.com/seasalt-ai/snowboy
why-ignite - Why should we use PyTorch-Ignite ?
vosk-api - Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
ignite - High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
Spoken-Keyword-Spotting - In this repository, we explore using a hybrid system consisting of a Convolutional Neural Network and a Support Vector Machine for Keyword Spotting task.
ompi - Open MPI main development repository
localcroft - Bits for locally-served Mycroft instances
gloo - Collective communications library with various primitives for multi-machine training.
C_to_Python_translator - Using File I/O we were able to convert C code written in one text file to Python code in another text file with the application of multiple function that could identify and accordingly process specific key words and formats used in the C language.