chaiko

torchtitan

A native PyTorch Library for large model training

xformers

Hackable and optimized Transformers building blocks, supporting a composable construction.

grok-1

Grok open release

maxtext

A simple, performant and scalable Jax LLM!

minbpe

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

llama

Inference code for Llama models

axlearn

An Extensible Deep Learning Library

torch2jax

Run PyTorch in JAX. 🤝

slurm

Slurm: A Highly Scalable Workload Manager

gost

GO Simple Tunnel - a simple tunnel written in golang

LxgwWenKai

An open-source Chinese font derived from Fontworks' Klee One. 一款开源中文字体，基于 FONTWORKS 出品字体 Klee One 衍生。

xg2xg

by ex-googlers, for ex-googlers - a lookup table of similar tech & services

stable-diffusion-webui-colab

stable diffusion webui colab

ml-stable-diffusion

Stable Diffusion with Core ML on Apple Silicon

speechmetrics

A wrapper around speech quality metrics MOSNet, BSSEval, STOI, PESQ, SRMR, SISDR

saxml

NeMo-text-processing

NeMo text processing for ASR and TTS

Audio_Stuff

Notes about my experience in audio industry

tuning_playbook

A playbook for systematically maximizing the performance of deep learning models.

pyroomacoustics

Pyroomacoustics is a package for audio signal processing for indoor applications. It was developed as a fast prototyping platform for beamforming algorithms in indoor scenarios.

torch-audiomentations

Fast audio data augmentation in PyTorch. Inspired by audiomentations. Useful for deep learning.

SPTK

A suite of speech signal processing tools

awesome-diarization

A curated list of awesome Speaker Diarization papers, libraries, datasets, and other resources.

cmudict

CMU US English Dictionary

dataloader

The merlin dataloader lets you rapidly load tabular data for training deep leaning models with TensorFlow, PyTorch or JAX

fiddle

wavesurfer.js

Audio waveform player

chinese_text_normalization

Chinese text normalization for speech processing

kernl

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

DeepFilterNet

Noise supression using deep filtering