VoiceAgent Library

Overview

The VoiceAgent library provides a set of components, hooks, and utilities for integrating AI-powered voice assistants into applications. It supports real-time messaging, audio device management, and connection handling for seamless interaction.

Installation

To install the package, run:

npm install rapida-react@latest

or using Yarn:

yarn add rapida-react@latest

Usage

Import the necessary components and hooks from rapida-react@latest:

import {
  Channel,
  ConnectionState,
  MediaDeviceFailure,
  ConnectionConfig,
  AgentConfig,
} from "rapida-react";

import {
  BrandIcon,
  UserIcon,
  MessagingAction,
  ConversationMessages,
  VoiceAgent,
  VoiceAgentContext,
  useConnectAgent,
  useDisconnectAgent,
  useInputModeToggleAgent,
  useMicInputToggleAgent,
  useSpeakerOuputToggleAgent,
  useMultiband3DSpeakerTrackVolume,
  useMultibandMicrophoneTrackVolume,
  useSelectInputDeviceAgent,
} from "rapida-react";

Example

Here is an example of how to use VoiceAgentContext.Provider to set up a voice agent:

<VoiceAgentContext.Provider
  value={
    new VoiceAgent(
      new ConnectionConfig({
        authorization: YOUR_TOKEN,
      }),
      new AgentConfig(ASSISTANT_ID).addArgument("key", "val")
    )
  }
>
  <ConversationMessages
    userIcon={UserIcon}
    brandIcon={BrandIcon}
    className="h-full overflow-auto !pb-40 no-scrollbar"
    intialConversations={[
      {
        role: "system",
        messages: [
          "I am Yuuki",
          "An AI-powered coach to help you master the toughest challenges at your workplace.",
        ],
      },
    ]}
  />
  <div className="absolute bottom-0 right-0 left-0">
    <MessagingAction
      className="h-[8rem]"
      placeholder="How can I help you overcome a challenge?"
    />
  </div>
</VoiceAgentContext.Provider>

Available Exports

The following components, hooks, and utilities are available for import:

Types & Configurations

• Channel: Defines communication channels.
• ConnectionState: Represents the state of a voice agent connection.
• MediaDeviceFailure: Handles media device errors.
• ConnectionConfig: Configures the connection settings for a voice agent.
• AgentConfig: Configures the agent settings, including parameters and keywords.

Components

• BrandIcon: Displays the brand's representative icon.
• UserIcon: Displays the user's representative icon.
• MessagingAction: A messaging input component for user interactions.
• ConversationMessages: Manages and displays the conversation flow.

Context

• VoiceAgent: The core AI voice agent instance.
• VoiceAgentContext: Provides the voice agent's context for managing interactions.

Hooks

• useConnectAgent: Establishes a connection to the voice agent.
• useDisconnectAgent: Disconnects the agent from the session.
• useInputModeToggleAgent: Toggles between different input modes.
• useMicInputToggleAgent: Enables or disables microphone input.
• useSpeakerOuputToggleAgent: Enables or disables speaker output.
• useMultiband3DSpeakerTrackVolume: Adjusts multiband volume for speaker tracks.
• useMultibandMicrophoneTrackVolume: Adjusts multiband volume for microphone tracks.
• useSelectInputDeviceAgent: Selects and manages input audio devices.

License

This project is licensed under the MIT License.