Awesome GitHub RepositoriesCross-Modal Alignment Models

GPT-SoVITS is a text-to-speech synthesis engine and voice cloning toolkit designed for generating natural-sounding human speech. It functions as a neural audio processing pipeline that maps input text to high-fidelity audio waveforms, utilizing conditional variational autoencoders and flow-based decoders to ensure expressive output. The platform distinguishes itself through its ability to perform few-shot voice cloning and cross-lingual speech generation, allowing users to maintain a specific speaker's vocal identity and emotional delivery across multiple languages. By employing cross-modal l

CosyVoice is a speech synthesis framework that utilizes large language models to generate expressive, multilingual audio. The system functions as an audio generation engine capable of producing natural-sounding speech across multiple languages while preserving regional dialects and specific emotional tones. The platform distinguishes itself through its zero-shot voice cloning capabilities, which allow for the creation of synthetic voice profiles from short audio samples without requiring additional model training. It provides fine-grained control over vocal attributes, enabling users to adjus

Dia is a generative AI audio tool and text-to-speech synthesis engine designed for the production-ready deployment of machine learning models. It provides a framework for creating lifelike synthetic speech by conditioning generation on reference audio samples to replicate specific vocal characteristics, emotional tones, and delivery styles. The system distinguishes itself through its ability to perform custom voice cloning and precise control over audio output. Users can adjust generation parameters such as temperature and guidance scale to modify the pacing, creativity, and style of the synt

This project is a self-supervised vision foundation model based on a vision transformer architecture. It is designed to learn dense visual representations from unlabeled images, serving as a general-purpose backbone for a wide variety of downstream vision tasks. The system is distinguished by its use of self-distillation and masked image modeling to extract semantic and geometric features. It also incorporates an image-text alignment model that maps visual embeddings to textual descriptions, enabling zero-shot image recognition, zero-shot segmentation, and cross-modal retrieval. The project

Point-e is a system for 3D model synthesis that generates three-dimensional point clouds from natural language descriptions and two-dimensional images. It utilizes diffusion models to synthesize these spatial representations based on text prompts or source images. The project includes specialized tools for refining these outputs, such as a point cloud upsampler to increase the density and resolution of low-resolution models. It also provides a mesh converter that uses distance function regression to transform raw point cloud data into structured 3D meshes. The broader capability surface cove

ml-mgie is a multimodal machine learning framework and image editor designed for instruction-based image manipulation. It utilizes multimodal large language models to translate natural language prompts into precise visual modifications, functioning as a text-to-image editing model. The system is a research implementation focused on aligning visual imagination with textual commands. It employs a training process based on image-pair datasets and descriptive instructions to learn how to execute complex visual edits. The framework covers capabilities in AI-powered visual content creation, includ

GLIDE is a generative model designed for text-to-image synthesis, image editing, and the contextual filling of masked image regions. It uses a guided diffusion process to transform random noise into high-resolution imagery that aligns with descriptive text prompts. The system provides specialized capabilities for modifying existing visuals, including the ability to alter specific image elements and iteratively refine selected regions through text-driven guidance. It also functions as an inpainting tool, filling missing or masked sections of an image with new content that blends naturally with

Recognize-anything is a multimodal foundation model designed for image recognition, visual tagging, and the generation of descriptive text captions from visual input. It functions as a multimodal embedding model that maps images and text into a shared vector space to enable cross-modal retrieval and recognition. The system implements zero-shot image classification and open-vocabulary object detection, allowing it to recognize object categories not present in the original training data through custom label embeddings. It also features a visual tagging engine and a captioning system that produc

Bark Voice Cloning est un moteur de synthèse texte-parole conçu pour générer un audio au son naturel et répliquer des caractéristiques vocales spécifiques. Le système utilise un modèle autorégressif basé sur transformer pour convertir le texte écrit en parole haute fidélité, prenant en charge la sortie multilingue et une livraison expressive. Le projet se distingue par le clonage de voix zero-shot, qui extrait les embeddings d'identité du locuteur à partir de courts échantillons audio pour conditionner le modèle génératif sans nécessiter de réglage fin étendu. Il fournit également des flux de travail spécialisés pour la conversion d'identité vocale, permettant aux utilisateurs de transformer le locuteur d'un enregistrement existant tout en préservant la livraison émotionnelle et les modèles rythmiques originaux. La plateforme englobe une suite complète d'outils pour la synthèse vocale et la manipulation audio. Cela inclut des utilitaires pour extraire l'audio source des médias, entraîner des modèles vocaux personnalisés et mapper le contenu linguistique sémantique vers des jetons acoustiques fins. Le logiciel est distribué sous forme d'une collection de notebooks Jupyter qui facilitent l'exécution de ces pipelines d'inférence multi-étapes.