Skip to main content

Talkscriber Text-to-Speech WebSocket API Documentation

Welcome to the Talkscriber Text-to-Speech WebSocket API documentation. This API is designed to provide seamless, real-time text-to-speech conversion using a secure WebSocket connection. Whether you're developing a TTS application or integrating speech synthesis capabilities into your platform, this documentation will guide you through the process.

Key Features​

  • Ultra-Low Latency Streaming: Speech starts in less than 0.1 seconds with real-time audio chunk processing
  • Real-time Audio Playback: Get immediate audio output as text is processed
  • Configurable Buffering: Adjustable buffer size for optimal latency vs. quality balance
  • Multiple Output Options: Support for real-time playback, file saving, or both
  • Cross-Platform: Works on Windows, macOS, and Linux
  • Secure Connection: Uses Secure WebSocket (WSS) for secure communication

Getting Started​

To start using the TTS API, connect to our test WebSocket URL:

wss://api.talkscriber.com:9099

For detailed instructions on setting up and using the API, follow the links in the Quick Links section.

Supported Audio Formats​

  • Sample Rate: 24kHz (matches server configuration)
  • Channels: Mono (1 channel)
  • Bit Depth: 16-bit PCM
  • Protocol: WebSocket binary streaming