Skip to main content

Getting Started

Introduction​

Whisperly Live is a real-time transcription system designed for continuous audio streaming. Leveraging WebSocket technology, it delivers low-latency transcription suitable for live events, meetings, and other applications requiring immediate feedback.

How It Works​

  1. Establish a WebSocket Connection:
    The client opens a persistent WebSocket connection to the server.
  2. Stream Audio Data:
    Audio is streamed in real-time to the server for transcription.
  3. Real-Time Transcription:
    The server processes the audio stream and continuously returns transcription segments, providing a live transcript as the audio is processed.

Features​

  • Real-Time Streaming:
    Offers continuous, low-latency transcription using WebSockets with GPU support.
  • Interactive Feedback:
    Receives transcription updates immediately during the session.
  • Scalable Architecture:
    Ideal for high-demand live events and real-time applications.
  • Secure Access:
    Uses API key-based authentication to ensure secure operations. You charged based on amount of duration_minutes.