ParseBench - A Document Parsing Benchmark for AI Agents

llama.cpp fork with additional SOTA quants and improved performance

Focus: a minimalist presentation theme for LaTeX Beamer.

A Beamer colour theme that maximizes visibility in dark and unfavourable conditions

Moloch is a minimalist, feature-rich Beamer theme for LaTeX presentations with a clean design and extensive customization options.

Simple, typographic beamer theme

A Python library to inspect and modify the internal structure of a PDF file

Get your documents ready for gen AI

A Faster LayoutReader Model based on LayoutLMv3, Sort OCR bboxes to reading order.

Baseline Recipe for VoicePrivacy Challenge 2024: anonymization systems and evaluation software

Confidence interval computation for evaluation in machine learning using the bootstrapping approach

Code for reproducting the paper Music Augmentation and Denoising For Peak-Based Audio Fingerprinting

Audio recorder plugin for Flutter with multiple options.

Code & dataset for the paper 'Attention-based Neural Text Segmentation'

The official repository of Dynamic-SUPERB.

Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context

VoicePAT is a modular and efficient toolkit for voice privacy research, with main focus on speaker anonymization.

SpeechGLUE is a speech version of the GLUE benchmark, driven by text-to-speech.

Evaluation results for Machine Translation within the BigScience project

Compare neural networks by their feature similarity

chinese speech pretrained models

Burj Khalifa Clustering method

Convert Machine Learning Code Between Frameworks

Inference code for LLaMA 2 models

Cast macOS and Linux Audio/Video to your Google Cast and Sonos Devices

A multilingual ASR model that can recognize ten Turkic languages—Azerbaijani, Bashkir, Chuvash, Kazakh, Kyrgyz, Sakha, Tatar, Turkish, Uyghur, and Uzbek.

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