Soul AI Lab ORG

[ACL 2026 Main] Training, inference, and testing of the SAC speech codec model.

SoulX-Podcast is an inference codebase by the Soul AI team for generating high-fidelity podcasts from text.

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.

Plug-and-play streaming semantic VAD for real-time full-duplex spoken dialogue systems.

Official inference code for SoulX-LiveAct: Towards Hour-Scale Real-Time Human Animation with Neighbor Forcing and ConvKV Memory

SoulX-FlashHead: A unified 1.3B-parameter framework designed for high-fidelity, infinite-length, and real-time streaming portrait video generation.

Official inference code for SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis

An end-to-end framework for multi-speaker transcription that jointly models who spoke, when, and what.

A Benchmark and Evaluation Suite for Zero-shot Singing Voice Synthesis

No description.

A high-throughput and memory-efficient inference and serving engine for LLMs