Python

In this section, we describe how to install the Python package sherpa-onnx.

• Install the Python Package
  • Method 1 (From pre-compiled wheels, CPU only)
  • Method 2 (From pre-compiled wheels, CPU + CUDA)
  • Method 3 (From source)
  • Method 4 (For developers)
  • Check your installation
• Decode files
  • Streaming zipformer
  • Non-streaming zipformer
  • Non-streaming paraformer
• Real-time speech recognition from a microphone
  • With endpoint detection
  • Without endpoint detection
• Speech recognition from URLs
  • Decode a URL
  • RTMP
    • Install the server
    • Start the server
    • Start ffmpeg to push audio stream
    • Start sherpa-onnx to pull audio stream
• Streaming WebSocket Server
  • Start the server
    • Use Python API
      • Send a file for decoding
      • Send audio samples from a microphone for decoding
    • Use a browser
  • colab
• Non-Streaming WebSocket Server
  • Non-streaming transducer
    • Start the server
    • Start the client
  • Non-streaming paraformer
    • Start the server
    • Start the client
  • Non-streaming CTC model from NeMo
    • Start the server
    • Start the client
  • Non-streaming Whisper tiny.en
    • Start the server
    • Start the client
  • colab