Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities。

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.

VITS2: Improving Quality and Efficiency of Single-Stage Text-to-Speech with Adversarial Learning and Architecture Design

Instant voice cloning by MIT and MyShell.

Inference and training library for high-quality TTS models.

AI powered speech denoising and enhancement. Adapted for windows and optimized

Noise supression using deep filtering

Multiple NVIDIA GPUs or Apple Silicon for Large Language Model Inference?

Upsampling Artifacts in Neural Audio Synthesis – https://arxiv.org/abs/2010.14356

End-to-End Zero-Shot Voice Conversion with Location-Variable Convolutions

🔊 Create labeled datasets, enhance audio quality, identify speakers, support diverse dataset types. 🎧👥📊 Advanced audio processing.

The official implementation of HierSpeech++

Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

AI powered speech denoising and enhancement

Faster Tortoise inference then Tortoise Fast Fork

Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Models

Flutter tools for Tizen

A rough and ready Python utility which splits audio files based on silence and desired min/max chunk duration.

Versatile audio super resolution (any -> 48kHz) with AudioSR.

Public voice datasets used for our Text-to-Speech voices.

TorToiSe fine-tuning with DLAS

Adaptive Vocoder for Custom Voice

Emote Portrait Alive: Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions

A multi-voice TTS system trained with an emphasis on quality

[SIGGRAPH Asia 2022] VideoReTalking: Audio-based Lip Synchronization for Talking Head Video Editing In the Wild

🔥 2D and 3D Face alignment library build using pytorch

The source code of "DINet: deformation inpainting network for realistic face visually dubbing on high resolution video."