Phonic Realtime Adapter for RTP Agent

The Phonic adapter provides seamless integration with Phonic's Multimodal Realtime API, allowing your RTP Agent to process voice and text natively with ultra-low latency.

Installation

Add the adapter to your Go project:

go get github.com/cavos-io/rtp-agent/adapter/phonic

Authentication

Set the required environment variables in your .env file. Refer to the Phonic developer documentation for acquiring the necessary API keys and tokens.

PHONIC_API_KEY=your_api_key_here

Usage

Below is a basic conceptual example demonstrating how to initialize the Phonic Realtime adapter within an RTP Agent session:

package main

import (
	"context"
	"log"
	"os"

	"github.com/cavos-io/rtp-agent/adapter/phonic"
	"github.com/cavos-io/rtp-agent/core/agent"
)

func main() {
	ctx := context.Background()

	// Initialize the Phonic Realtime adapter
	realtimeProvider, err := phonic.NewRealtimeModel(
		os.Getenv("PHONIC_API_KEY"),
	)
	if err != nil {
		log.Fatalf("failed to initialize Phonic adapter: %v", err)
	}

	// Create and configure the RTP agent session
	session := agent.NewMultimodalSession(
		agent.WithRealtimeModel(realtimeProvider),
	)

	// Start the session
	if err := session.Start(ctx); err != nil {
		log.Fatalf("session failed: %v", err)
	}
}