Skip to content

@Soul-AILab Soul AI Lab

Change the repository type filter

All

    Repositories list

    11 repositories

    • SoulX-FlashTalk

      Public
      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.
      Python
      •
      Apache License 2.0
      •137•1.5k•16•4•Updated Jul 30, 2026Jul 30, 2026

    • SoulX-Duplug

      Public
      Plug-and-play streaming semantic VAD for real-time full-duplex spoken dialogue systems.
      streamingfull-duplexspoken-language-understanding
      streamingfull-duplexspoken-language-understandingspoken-dialogue-systemssemantic-vad
      Python
      •
      Apache License 2.0
      •39•295•12•0•Updated Jul 17, 2026Jul 17, 2026

    • SoulX-Transcriber

      Public
      An end-to-end framework for multi-speaker transcription that jointly models who spoke, when, and what.
      speech-recognitionsdrasr
      speech-recognitionsdrasrsdllm
      Python
      •
      Apache License 2.0
      •15•289•9•0•Updated Jun 22, 2026Jun 22, 2026

    • SoulX-LiveAct

      Public
      Official inference code for SoulX-LiveAct: Towards Hour-Scale Real-Time Human Animation with Neighbor Forcing and ConvKV Memory
      Python
      •114•1.2k•14•4•Updated Jun 15, 2026Jun 15, 2026

    • Soul-AILab.github.io

      Public
      JavaScript
      •5•17•0•0•Updated Jun 2, 2026Jun 2, 2026

    • SoulX-Singer

      Public
      Official inference code for SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis
      Python
      •
      Apache License 2.0
      •142•917•31•1•Updated May 29, 2026May 29, 2026

    • SoulX-FlashHead

      Public
      SoulX-FlashHead: A unified 1.3B-parameter framework designed for high-fidelity, infinite-length, and real-time streaming portrait video generation.
      Python
      •
      Apache License 2.0
      •116•989•27•1•Updated May 28, 2026May 28, 2026

    • SoulX-Singer-Eval

      Public
      A Benchmark and Evaluation Suite for Zero-shot Singing Voice Synthesis
      Python
      •
      Apache License 2.0
      •3•34•0•0•Updated Feb 11, 2026Feb 11, 2026

    • SoulX-Podcast

      Public
      SoulX-Podcast is an inference codebase by the Soul AI team for generating high-fidelity podcasts from text.
      Python
      •
      Apache License 2.0
      •459•3.5k•40•9•Updated Dec 11, 2025Dec 11, 2025

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •21k•1•0•0•Updated Nov 3, 2025Nov 3, 2025

    • SAC

      Public
      [ACL 2026 Main] Training, inference, and testing of the SAC speech codec model.
      semanticcodecacoustic
      semanticcodecacousticspeech-reconstructionspeech-representationspeech-disentanglement
      Python
      •
      Apache License 2.0
      •8•109•0•0•Updated Nov 1, 2025Nov 1, 2025
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.