SoulX-FlashTalk
SoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.
About
SoulX-FlashTalk is a software tool designed for real-time infinite streaming of audio-driven avatars. It utilizes self-correcting bidirectional distillation to achieve this functionality, making it suitable for applications in virtual communication and entertainment. The product is built using Python and aims to provide a seamless experience for users looking to integrate animated avatars into their audio interactions.
Open Source Health
- Stars
- 1,507
- Forks
- 140
- License
- Apache-2.0
- Last commit
- 2 months ago
Resources & Links
Related Categories
Vendor
Soul AI Lab
Open-source project on GitHub: SoulX-FlashHead
Quick Links
Open Source
More by Soul AI Lab
Related Products
CHaiDNN
HLS based Deep Neural Network Accelerator Library for Xilinx Ultrascale+ MPSoCs
Top category match
Ts Pattern
The exhaustive Pattern Matching library for TypeScript, with smart type inference.
Top category match
Nobodywho
NobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.
Top category match
ds4
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
Top category match
RECCON
This repository contains the dataset and the PyTorch implementations of the models from the paper Recognizing Emotion Cause in Conversations.
Top category match