集成音视频 Realtime iOS SDK

├── Podfile
├── coze-swift
│   ├── AppDelegate.swift
│   ├── Assets.xcassets
│   ├── Base.lproj
│   ├── Config                        //配置 API 相关信息
│   ├── Info.plist
│   ├── Models                       //创建房间相关数据结构
│   ├── SceneDelegate.swift
│   ├── Services                       //API 请求封装
│   └── ViewController.swift           //主界面
├── coze-swift.xcodeproj
└── coze-swift.xcworkspace

├── Podfile
├── Pods
├── coze-objc
│   ├── AppDelegate.h
│   ├── AppDelegate.m
│   ├── Assets.xcassets
│   ├── Base.lproj
│   ├── Config
│   ├── Info.plist
│   ├── Models
│   ├── SceneDelegate.h
│   ├── SceneDelegate.m
│   ├── Services
│   ├── ViewController.h
│   ├── ViewController.m
│   └── main.m
├── coze-objc.xcodeproj
└── coze-objc.xcworkspace

import UIKit
import VolcEngineRTC

class ViewController: UIViewController, ByteRTCVideoDelegate, ByteRTCRoomDelegate {
    var rtcVideo: ByteRTCVideo?
    var rtcRoom: ByteRTCRoom?
    private var roomInfo: RoomData?
}

#import <UIKit/UIKit.h>
#import <VolcEngineRTC/VolcEngineRTC.h>
#import "ApiResponse.h"

NS_ASSUME_NONNULL_BEGIN

@interface ViewController : UIViewController <ByteRTCVideoDelegate, ByteRTCRoomDelegate, UITableViewDelegate, UITableViewDataSource>
@end

NS_ASSUME_NONNULL_END

func createUI() {
    let width = self.view.bounds.size.width
    let height = self.view.bounds.size.height
    
    // 本地预览视图占上半部分
    let previewHeight = height * 0.4
    localView.frame = CGRect(x: 0, y: 0, width: width, height: previewHeight)
    self.view.addSubview(localView)
    
    // 按钮区域
    let buttonY = previewHeight + 10
    let buttonWidth = (width - 30) / 2
    
    joinButton.frame = CGRect(x: 10, y: buttonY, width: buttonWidth, height: 44)
    cameraButton.frame = CGRect(x: width - buttonWidth - 10, y: buttonY, width: buttonWidth, height: 44)
    
    self.view.addSubview(joinButton)
    self.view.addSubview(cameraButton)
}

- (void)createUI {
  CGFloat width = self.view.bounds.size.width;
  CGFloat height = self.view.bounds.size.height;

  // 本地预览视图
  CGFloat previewHeight = height * 0.4;
  self.localView = [[UIView alloc] initWithFrame:CGRectMake(0, 0, width, previewHeight)];
  self.localView.backgroundColor = [UIColor lightGrayColor];
  [self.view addSubview:self.localView];

  // 按钮区域
  CGFloat buttonY = previewHeight + 10;
  CGFloat buttonWidth = (width - 30) / 2;

  // 创建连接按钮
  self.joinButton = [UIButton buttonWithType:UIButtonTypeCustom];
  self.joinButton.frame = CGRectMake(10, buttonY, buttonWidth, 44);
  [self.joinButton setTitle:@"连接" forState:UIControlStateNormal];
  [self.joinButton addTarget:self action:@selector(connectButtonTapped) forControlEvents:UIControlEventTouchUpInside];
  
  // 创建其他控制按钮...
  
  [self.view addSubview:self.joinButton];
}

do {
    // 获取房间信息
    let response = try await NetworkService.shared.createRoom(
        botId: APIConfig.botId,
        voiceId: APIConfig.voiceId
    )
    
    // 异常处理
    if response.code != 0 {
        throw NSError(
            domain: "", code: Int(response.code),
            userInfo: [NSLocalizedDescriptionKey: response.msg])
    }
    roomInfo = response.data
} catch {
    print("连接失败: \(error)")
}

[[NetworkService shared] createRoomWithBotId:API_BOT_ID
                                   voiceId:API_VOICE_ID
                                completion:^(RoomResponse *response, NSError *error) {
    if (error || response.code != 0) {
        // 处理错误...
    } else {
        self.roomInfo = response.data;
        [self buildRTCEngine];
        [self bindLocalRenderView];
        [self joinRoom];
    }
}];

func buildRTCEngine() {
    guard let roomInfo = self.roomInfo else { return }
    
    self.rtcVideo = ByteRTCVideo.createRTCVideo(
        roomInfo.app_id, delegate: self, parameters: [:])
}

- (void)buildRTCEngine {
  if (!self.roomInfo) {
    return;
  }

  self.rtcVideo = [ByteRTCVideo createRTCVideo:self.roomInfo.app_id
                                    delegate:self
                                  parameters:@{}];
  [self.rtcVideo startAudioCapture];
}

func bindLocalRenderView() {
    let canvas = ByteRTCVideoCanvas.init()
    canvas.view = self.localView
    canvas.renderMode = .hidden
    self.rtcVideo?.setLocalVideoCanvas(.indexMain, withCanvas: canvas)
}

- (void)bindLocalRenderView {
  ByteRTCVideoCanvas *canvas = [[ByteRTCVideoCanvas alloc] init];
  canvas.view = self.localView;
  canvas.renderMode = ByteRTCRenderModeHidden;
  [self.rtcVideo setLocalVideoCanvas:ByteRTCStreamIndexMain withCanvas:canvas];
}

func addMessage(_ message: String, eventType: String) {
    DispatchQueue.main.async {
        if self.lastEventType == "conversation.message.delta" {
            if var lastMessage = self.messageList.last {
                lastMessage += message
                self.messageList[self.messageList.count - 1] = lastMessage
            }
        } else if eventType == "conversation.message.delta" {
            self.messageList.append(message)
        }
        
        self.lastEventType = eventType
        self.messageTableView.reloadData()
    }
}

- (void)rtcRoom:(ByteRTCRoom *)rtcRoom
    onUserMessageReceived:(NSString *)uid
                  message:(NSString *)message {
  NSError *error;
  NSData *jsonData = [message dataUsingEncoding:NSUTF8StringEncoding];
  NSDictionary *messageDict = [NSJSONSerialization JSONObjectWithData:jsonData
                                                            options:0
                                                              error:&error];

  if (!error) {
    NSString *eventType = messageDict[@"event_type"];
    if ([eventType isEqualToString:@"conversation.message.delta"]) {
      NSString *content = messageDict[@"data"][@"content"] ?: @"";
      [self addMessage:content eventType:eventType];
    }
  }
}

@objc private func cameraButtonTapped() {
    cameraButton.isSelected = !cameraButton.isSelected
    
    if cameraButton.isSelected {
        self.rtcVideo?.startVideoCapture()
    } else {
        self.rtcVideo?.stopVideoCapture()
    }
}

- (void)cameraButtonTapped {
  self.cameraButton.selected = !self.cameraButton.selected;

  if (self.cameraButton.selected) {
    [self.rtcVideo startVideoCapture];
  } else {
    [self.rtcVideo stopVideoCapture];
  }
}

@objc private func muteButtonTapped() {
    muteButton.isSelected = !muteButton.isSelected
    
    if muteButton.isSelected {
        self.rtcVideo?.stopAudioCapture()
    } else {
        self.rtcVideo?.startAudioCapture()
    }
}

- (void)muteButtonTapped {
  self.muteButton.selected = !self.muteButton.selected;

  if (self.muteButton.selected) {
    [self.rtcVideo stopAudioCapture];
  } else {
    [self.rtcVideo startAudioCapture];
  }
}

@objc private func interruptButtonTapped() {
    let message = [
        "id": "event_1",
        "event_type": "conversation.chat.cancel",
        "data": "{}"
    ]
    
    do {
        let jsonData = try JSONSerialization.data(withJSONObject: message)
        if let jsonString = String(data: jsonData, encoding: .utf8) {
            self.rtcRoom?.sendUserMessage(
                APIConfig.botId, 
                message: jsonString,
                config: ByteRTCMessageConfig.reliableOrdered)
        }
    } catch {
        print("Error creating JSON: \(error)")
    }
}

- (void)interruptButtonTapped {
  NSDictionary *message = @{
    @"id" : @"event_1",
    @"event_type" : @"conversation.chat.cancel",
    @"data" : @"{}"
  };

  NSError *error;
  NSData *jsonData = [NSJSONSerialization dataWithJSONObject:message
                                                   options:0
                                                     error:&error];

  if (!error) {
    NSString *jsonString = [[NSString alloc] initWithData:jsonData
                                               encoding:NSUTF8StringEncoding];
    [self.rtcRoom sendUserMessage:API_BOT_ID
                        message:jsonString
                         config:ByteRTCMessageConfigReliableOrdered];
  }
}

deinit {
    // 销毁房间
    self.rtcRoom?.leaveRoom()
    self.rtcRoom?.destroy()
    self.rtcRoom = nil
    
    // 销毁引擎
    ByteRTCVideo.destroyRTCVideo()
    self.rtcVideo = nil
}

- (void)dealloc {
  [self.rtcRoom leaveRoom];
  [self.rtcRoom destroy];
  self.rtcRoom = nil;
  [ByteRTCVideo destroyRTCVideo];
  self.rtcVideo = nil;
}

// 发送消息
let message = [
    "id": "event_1",
    "event_type": "conversation.message.create",
    "data": [
        "role": "user",
        "content_type": "text",
        "content": "讲一个笑话"
    ],
]

do {
    let jsonData = try JSONSerialization.data(withJSONObject: message)
    if let jsonString = String(data: jsonData, encoding: .utf8) {
        self.rtcRoom?.sendUserMessage(
            APIConfig.botId, message: jsonString,
            config: ByteRTCMessageConfig.reliableOrdered)
    }
} catch {
    print("Error creating JSON: \(error)")
}

// 发送消息
  NSDictionary *message = @{
    @"id" : @"event_1",
    @"event_type" : @"conversation.message.create",
    @"data" : @{
        @"role": @"user",
        @"content_type": @"text",
        @"content": @"讲一个笑话"
    }
  };

  NSError *error;
  NSData *jsonData = [NSJSONSerialization dataWithJSONObject:message
                                                     options:0
                                                       error:&error];

  if (!error) {
    NSString *jsonString = [[NSString alloc] initWithData:jsonData
                                                 encoding:NSUTF8StringEncoding];
    [self.rtcRoom sendUserMessage:API_BOT_ID
                          message:jsonString
                           config:ByteRTCMessageConfigReliableOrdered];
  }

func rtcRoom(_ rtcRoom: ByteRTCRoom, onUserMessageReceived uid: String, message: String) {
    print("收到用户消息 - 用户ID: \(uid), 消息: \(message)")

    do {
        if let jsonData = message.data(using: .utf8) {
            let messageData = try JSONDecoder().decode(MessageData.self, from: jsonData)

            if messageData.event_type == "conversation.message.delta"
                || messageData.event_type == "conversation.message.completed"
            {
                let content = messageData.data?.content ?? ""
                self.addMessage(content, eventType: (messageData.event_type)!)
            }
        }
    } catch {
        print("JSON 解析错误: \(error)")
    }
}

// 监听所有事件
- (void)rtcRoom:(ByteRTCRoom *)rtcRoom
    onUserMessageReceived:(NSString *)uid
                  message:(NSString *)message {
  NSLog(@"收到用户消息 - 用户ID: %@, 消息: %@", uid, message);

  NSError *error;
  NSData *jsonData = [message dataUsingEncoding:NSUTF8StringEncoding];
  NSDictionary *messageDict = [NSJSONSerialization JSONObjectWithData:jsonData
                                                              options:0
                                                                error:&error];
}