proger

fsevent_watch

simple but awesome OS X FSEvents client

gentle-uk

Лагідна Українізація

nsf

Neural Source Filter

opus-vad

Use modern audio compression for the internet for voice activity segmentation.

gecbot

mova122

Через рік стане краще

nanoGPT

The simplest, fastest repository for training/finetuning medium-sized GPTs.

r9

resnet9 for personal computing

sha-rnn

Single Headed Attention RNN - "Stop thinking with your head"

world-model

Discrete world modeling by recording Coppelia simulations with ROS

algebra

api_nlp_uk

b1

block-recurrent-transformer-pytorch

Implementation of Block Recurrent Transformer - Pytorch

BScNets

choppa

Partial python port of java SRX segmenter

datasette-audio

Turn .mp3/wav URLs into an audio player in the Datasette interface

faster-whisper

Faster Whisper transcription with CTranslate2

flash-attention

Fast and memory-efficient exact attention

icefall

iris

Transformers are Sample-Efficient World Models. ICLR 2023, notable top 5%.

kaldi

kaldi-asr/kaldi is the official location of the Kaldi project.

nlp_uk_api

ogbl_d

omnigres

Postgres as a Platform

quantorch

sherpa

Streaming and non-streaming ASR server for next-gen Kaldi

ukraine_itn_wfst

Simple WFST for Ukrainian ITN based on NVIDIA NeMo and Pynini

vae-bow

Bernoulli VAE for bags of words

whisper

Robust Speech Recognition via Large-Scale Weak Supervision