Notifications

0 unread notifications

No unread notifications

You're all caught up!

Libraries And Sdk

@ageofai/vp-ui

A beautiful, reusable voice interface component for React applications. Features real-time audio visualization, transcript display, and seamless integration with Pipecat AI servers.

Real-time Voice Interface
Audio Visualization
Live Transcription

Basic Configuration

Configure your voice interface with essential connection settings, authentication tokens, and basic features like microphone and camera access.

Key Properties:

  • serverUrl - Pipecat server endpoint
  • systemPrompt - AI assistant instructions
  • authToken - Authentication token
  • enableMic/enableCam - Media permissions

Theme & Styling

Customize the appearance with built-in themes, size variants, and control which UI elements are visible to match your application's design.

Available Options:

  • theme - 'dark' | 'light'
  • size - 'sm' | 'md' | 'lg'
  • showInfo/showShare/showTranscript - UI toggles

Advanced Styling

Apply custom CSS classes to different parts of the component and add custom event handlers for sharing and info actions.

Custom Classes:

  • className - Main container
  • orbClassName - Voice orb styling
  • controlsClassName - Control buttons
  • transcriptClassName - Transcript area

Using the Hook Separately

For advanced use cases, use the useAgeOfAiConnect hook independently to build completely custom voice interfaces.

Hook Returns:

  • client - WebSocket client instance
  • isConnected/isConnecting - Connection state
  • transcript - Message history
  • isBotSpeaking/isUserSpeaking - Audio state

@ageofai/vp-sdk

The official Node.js SDK for the Scoreexl API—a secure audio AI platform providing Speech-to-Text (STT), Text-to-Speech (TTS) with Google Chirp3 HD & Deepgram Aura-2, RAG-based knowledge, analytics, credits, and notifications.

Complete API Coverage
Analytics & Credits
TypeScript Support

Authentication

Manage user accounts with registration, login, and logout functionality. All API calls require authentication.

Available Methods:

  • register() - Create new user account
  • login() - Authenticate existing user
  • logout() - End user session

Text-to-Speech (TTS)

Convert text to high-quality speech using Google Chirp3 HD and Deepgram Aura-2 voices. Support for multiple languages and voice styles.

TTS Workflow:

  • 1. List available voices
  • 2. Submit synthesis request
  • 3. Poll for completion status
  • 4. Download generated audio

Speech-to-Text (STT)

Transcribe audio files to text with high accuracy. Supports multiple audio formats and provides detailed segment information with timestamps.

STT Features:

  • • Multiple audio format support
  • • Detailed segment timestamps
  • • High accuracy transcription
  • • Async processing with status polling

Credits & Analytics

Monitor your API usage, manage credits, and access detailed analytics. Track consumption patterns and optimize your usage.

Management Features:

  • • Real-time credit balance
  • • Usage analytics and reporting
  • • Credit purchase system
  • • Historical usage data

RAG & Agents

Leverage Retrieval-Augmented Generation (RAG) for document-based Q&A and run specialized AI agents for various tasks.

Advanced AI Features:

  • • Document-based RAG sessions
  • • Specialized AI agents
  • • Context-aware responses
  • • Multi-document querying

Minimal Voice UI

A clean, minimal voice interface setup perfect for simple applications or prototyping.

Next.js Integration

Full-featured Next.js page with environment variables, custom styling, and responsive design.

Complete TTS Workflow

End-to-end text-to-speech implementation with authentication, voice selection, synthesis, and audio download.

STT with Error Handling

Robust speech-to-text implementation with comprehensive error handling and status polling.

TypeScript Support

Both the UI library and SDK provide comprehensive TypeScript definitions for better development experience and type safety.

Available Types:

  • VoiceUIProps - Component props
  • UseAgentConnectOptions - Hook options
  • TranscriptMessage - Message structure

Error Handling

All SDK errors extend a base ScoreexlError and include message, statusCode, code, and details properties for comprehensive error handling.

Error Types:

  • AuthenticationError - Login/auth issues
  • RateLimitError - API rate limiting
  • TranscriptionError - STT failures
  • SynthesisError - TTS failures

Requirements & Configuration

System requirements and advanced configuration options for both the UI library and SDK.

System Requirements:

  • React: 18.0.0 or higher
  • Node.js: 16.0.0 or higher
  • Tailwind CSS: For styling (recommended)
  • Pipecat Server: Running instance for voice processing