uis-rnn
lightning-bolts
Our great sponsors
uis-rnn | lightning-bolts | |
---|---|---|
3 | 3 | |
1,529 | 1,640 | |
0.3% | 1.4% | |
3.5 | 7.5 | |
8 months ago | 17 days ago | |
Python | Python | |
Apache License 2.0 | Apache License 2.0 |
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.
uis-rnn
-
[D] Is there a way to distinguish different human voices from 1 audio file ?
Looks like you can get an put of the box here: https://github.com/google/uis-rnn
-
Putting my degree to use. (Exclude Specials and Guests)
Discussion: - When I started this, I thought I would use something like the VoxSort Diarization and it would be easy. But these apps are terrible, especially in recognizing Joey apart from Garnt. Connor has a distinct voice so it was recognizable but still bad. But I didn't think Joey's and Garnt's voices were so similar. - Tested the thing and it's accuracy is almost 99%. - You can still improve this by cutting the episode into smaller chunk but 1 second is the maximum for my computer, any smaller than that i will run out of RAM. I can work to get around this but hey I'm lazy. - The library to implement yourself from google.
-
Finally, my degree can be useful
I used this algorithm from Google to determine "who spoke when".
lightning-bolts
-
Question about implementing RL algorithms
In the lightning-bolts repository, they implement the different RL algorithms, such as PPO and DQN, as different models. Would it make more sense to have the different algorithms be the Trainer instead? Inside each of the implementations, the model creates the same neural network with different training steps.
-
[P] An elegant and strong PyTorch Trainer
This is also how lightning bolts tend to be defined: for example, you have https://github.com/Lightning-AI/lightning-bolts/blob/master/pl_bolts/models/rl/advantage_actor_critic_model.py for A3C, which itself is only the loss and training wrapper around the normal modules for critic and actor (see https://github.com/Lightning-AI/lightning-bolts/blob/52e4c503c671f4866339c1537cf6ae506e7c5cf5/pl_bolts/models/rl/common/networks.py#L147=)
-
[D] How to organize deep learning projects on Github ?
Also PyTorch Lighting gives example in this repo https://github.com/PyTorchLightning/pytorch-lightning-bolts
What are some alternatives?
pyannote-audio - Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding
Kedro - Kedro is a toolbox for production-ready data science. It uses software engineering best practices to help you create data engineering and data science pipelines that are reproducible, maintainable, and modular.
pyDenStream - Implementation of the DenStream algorithm in Python.
detectron2 - Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.
orange - 🍊 :bar_chart: :bulb: Orange: Interactive data analysis
cookiecutter-pytorch - A Cookiecutter template for PyTorch Deep Learning projects.
hover - :speedboat: Label data at scale. Fun and precision included.
thinc - 🔮 A refreshing functional take on deep learning, compatible with your favorite libraries
ECAPA-TDNN - Unofficial reimplementation of ECAPA-TDNN for speaker recognition (EER=0.86 for Vox1_O when train only in Vox2)
haystack - :mag: LLM orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
Clover - An Efficient DNA Clustering algorithm based on Tree Structure.
PanelCleaner - An AI-powered tool to clean manga panels.