SOTA Open Source TTS

[arXiv 2024] Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis

PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation

Generative AI extensions for onnxruntime

Code for the paper Hybrid Spectrogram and Waveform Source Separation

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.

Official repository for LTX-Video

OCR software for recognition of handwritten text

OCR, layout analysis, reading order, table recognition in 90+ languages

DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

Reasoning in Large Language Models: Papers and Resources, including Chain-of-Thought and OpenAI o1 🍓

A library for advanced large language model reasoning

Ocular is a state-of-the-art historical OCR system.

Python-based tools for document analysis and OCR

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer

⚡ Build your chatbot within minutes on your favorite device; offer SOTA compression techniques for LLMs; run LLMs efficiently on Intel Platforms⚡

An innovative library for efficient LLM inference via low-bit quantization

Official inference framework for 1-bit LLMs

Low-bit LLM inference on CPU with lookup table

Fast inference engine for Transformer models

Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable…

Code for the paper Hybrid Spectrogram and Waveform Source Separation

Efficient vision foundation models for high-resolution generation and perception.

Port of OpenAI's Whisper model in C/C++

A throughput-oriented high-performance serving framework for LLMs

Manipulate audio with a simple and easy high level interface

extract text from any document. no muss. no fuss.

PDF to Markdown with vision models

Official code for "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"