AIAS (AI Accelerator Suite)

• AIAS provides capabilities:

1. SDK: includes support for various model hubs and GitHub preferred models.
2. Platform engine: includes API platform engine, search engine, training engine, edge computing engine, etc.
3. Scene suite: includes suites for various scenes such as ToB, ToC, and ToG, such as the biomedicine suite.

Project List:

• 1. image_sdks

1) Toolbox series: image processing toolbox (static image).
  2) Object detection: object detection, object tracking, face detection & recognition.
  3) Image segmentation: image segmentation, remote sensing images, medical images.
  4) Behavior analysis: behavior recognition, pose estimation.
  5) GAN: super resolution, motion-driven, style transfer, image generation.
  6) Other categories: OCR, SLAM, depth estimation, autonomous driving, reinforcement learning, video understanding, image fusion, image retrieval, etc.
      ...

• 2. nlp_sdks

  1) Toolbox series: Tokenizer, sentencepiece, fastText, npy/npz file processing, etc.
  2) Text generation
  3) Word vectors
  4) Machine translation
  5) Semantic model
  6) Emotional analysis
  7) Syntactic analysis
  8) Lexical analysis
  9) Text review
      ...

• 3. audio_sdks

  1) Toolbox series: phoneme toolbox, librosa, java sound, javacv ffmpeg, fft, vad toolbox, etc.
  2) Sound cloning
  3) Speech synthesis
  4) Voiceprint recognition
  5) Speech recognition
      ...

• 4. engine_hub

  1) Training engine
  2) Search engine
  3) API capability platform
      ...

• 5. suite_hub

1) ToB: question and answer system, etc.
  2) Biomedicine
  3) Digital virtual humans
  4) AI content generation (image generation: stable diffusion)
      ...

Other open source projects:

1. quantum computing

• GitHub: https://github.com/mymagicpower/qubits

2. AI + biomedicine

• GitHub: https://github.com/mymagicpower/bio-computing

3. AI desktop Apps

• GitHub: https://github.com/mymagicpower/easy_AI_apps