ChatGPT UI For SwiftUI

Simple and extensible Drop-in ChatGPT UI Solution for Apple Platforms.

Supported Platforms

• iOS 17
• macOS 14
• visionOS 1.x

Installation

Swift Package Manager

• File > Swift Packages > Add Package Dependency
• Copy and paste this URL

https://github.com/alfianlosari/ChatGPTUI.git

Import

Import to your project source file.

import ChatGPTUI

Requirement

Register for API key fromx OpenAI. Initialize with api key

Voice Chat

‼️ You need to add key of Privacy - Microphone Usage Description in info.plist when using this otherwise your app will crash ‼️

macOS Users

For App Sandbox, check these checkboxes in Xcode:

• Network Outgoing Connections (Client)
• Hardware Audio Input (For Voice Chat)
• Resource Access Audio Input (For Voice Chat)

Usage

Text Chat View

Simply initialize passing the apiKey in SwiftUI View

var body: some View {
    NavigationStack {
        TextChatView(apiKey: apiKey)
            .navigationTitle("XCA ChatGPTUI")
    }
}

Pass these optional params for customization:

• senderImage and botImage from asset or remote image URL to render as sender and bot images.
• model ChatGPTModel enum (gpt-4o, gpt-4turbo, gpt-3.5, etc)
• systemText system text prompt used by ChatGPT.
• temperature Temperature used by ChatGPT for response.

var body: some View {
    NavigationStack {
        TextChatView(
            senderImage: senderImage,
            botImage: botImage,
            model: .gpt_hyphen_4o,
            systemText: "You're master of Swift Programming",
            temperature: 1.0,
            apiKey: apiKey)
            .navigationTitle("XCA ChatGPTUI")
    }
}

Make sure you are in tier that eligible for gpt4 models access. You can learn more from here How can I access GPT-4, GPT-4 Turbo and GPT-4o?.

If you're not sure just pass gpt-3.5 models, by default it uses gpt-3.5turbo

Voice Chat View

Converse with ChatGPT using Voice. It uses 3 OpenAI APIs under the hood:

• Transcribe user speech to text.
• Prompt ChatGPT API using the text.
• Convert text to speech (TTS) using Whisper API.

‼️ You need to add key of Privacy - Microphone Usage Description in info.plist when using this otherwise your app will crash ‼️

Simply initialize passing the apiKey in SwiftUI View

var body: some View {
    NavigationStack {
        VoiceChatView(apiKey: apiKey)
            .navigationTitle("XCA ChatGPTUI")
    }
}

Pass these optional params for customization:

• Voice Type: select voice type from alloy (default), echo, fable, onyx, nova, shimmer
• model ChatGPTModel enum (gpt-4o, gpt-4turbo, gpt-3.5, etc)
• systemText system text prompt used by ChatGPT.
• temperature Temperature used by ChatGPT for response.

var body: some View {
    NavigationStack {
        VoiceChatView(
            voiceType: .nova,
            model: .gpt_hyphen_4o,
            systemText: "You're master of Swift Programming",
            temperature: 1.0,
            apiKey: apiKey)
            .navigationTitle("XCA ChatGPTUI")
        
    }
}