AWS: 媒介构造库

本单元中更高层次的构件的API是实验性的并正在积极开发中. 在今后任何版本中,它们都须作非背向相容的修改或去除。 这些是 不适用 语义版本 模型和断开的更改 在发行笔记中宣布. 这意味着虽然你可能使用它们,但可能需要更新 当更新到此软件包的更新版本时,您的源代码.

AWS 元素介质Live

AWS Elemental MediaLive是一种实时视频服务,可以让您为广播和流播制作现场节目.

这个软件包包含与 AWS Elemental MediaLive 合作的构造,包括输入,输入安全组,频道,以及MediaLive Where资源(网络,集群,频道定位组,SDI源).

关于AWS元素介质Live的进一步信息,参见: 文件。见 每个输出组中支持的代码。 。 。 .

以下示例创建了 SRT 调用器输入,将其编码为 H.264 + AAC,并将 HLS 段输出为 S3 桶 :

declare const stack: Stack;
declare const bucket: s3.IBucket;

const input = new medialive.Input(stack, 'SrtInput', {
  inputName: 'my-srt-input',
  input: medialive.InputConfiguration.srtCaller([{
    srtListenerAddress: '203.0.113.10',
    srtListenerPort: 5000,
  }]),
});

const video = medialive.EncodeConfiguration.video({
  name: 'video_720p',
  codec: medialive.VideoCodecSettings.h264({
    rateControl: medialive.H264RateControl.cbr({ bitrate: Bitrate.mbps(3) }),
    framerate: medialive.Framerate.FPS_30,
  }),
  width: 1280,
  height: 720,
});

const audio = medialive.EncodeConfiguration.audio({
  name: 'audio_aac',
  codec: medialive.AudioCodecSettings.aac({ bitrate: Bitrate.kbps(192) }),
});

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
    }),
  ],
});

投入

一个输入代表一个MediaLive信道的上游来源。 使用 InputConfiguration 制造不同输入类型的工厂方法.

SRT 呼叫器

MediaLive 连接到远程 SRT 听器 :

declare const stack: Stack;
new medialive.Input(stack, 'SrtInput', {
  inputName: 'srt-caller',
  input: medialive.InputConfiguration.srtCaller([{
    srtListenerAddress: '203.0.113.10',
    srtListenerPort: 5000,
  }]),
});

SRT 监听器

MediaLive 为即将到来的SRT连接收听. SRT 听器输入需要输入安全 组合。 要接收加密内容,请提供 a decryption 块引用密件管理器 秘密,持有密码句——秘密通过参考传递,所以MediaLive解决了ARN 合成时间 :

declare const stack: Stack;
declare const passphrase: secretsmanager.ISecret;

const sg = new medialive.InputSecurityGroup(stack, 'SrtSg', {
  allowlistRules: ['203.0.113.0/24'],
});

new medialive.Input(stack, 'SrtListenerInput', {
  inputName: 'srt-listener',
  input: medialive.InputConfiguration.srtListener({
    inputSecurityGroups: [sg],
    minimumLatency: Duration.millis(500),
    streamId: 'my-stream-id',
    decryption: {
      algorithm: medialive.SrtDecryptionAlgorithm.AES256,
      passphraseSecret: passphrase,
    },
  }),
});

AWS 元素介质连接路由器

创建自动加密的MediaConnect路由器输入 :

declare const stack: Stack;
new medialive.Input(stack, 'RouterInput', {
  inputName: 'mc-router',
  input: medialive.InputConfiguration.mediaConnectRouter(),
});

仅此一种输入 @aws-cdk/aws-mediaconnect-alpha是因为 RouterOutputConfiguration.mediaLiveInput() 可以交付——将其指向任何其他输入类型合成,但在部署时失败.

来自 S3 的 MP4 文件

使用 InputSource.fromBucket() 引用 S3 对象 :

declare const stack: Stack;
declare const bucket: s3.IBucket;

new medialive.Input(stack, 'FileInput', {
  inputName: 'mp4-file',
  input: medialive.InputConfiguration.mp4File([
    medialive.InputSource.fromBucket(bucket, 'media/input.mp4'),
  ]),
});

导入已有输入

declare const stack: Stack;
const input = medialive.Input.fromInputArn(stack, 'Imported',
  'arn:aws:medialive:us-east-1:123456789012:input:1234567');

输入安全组

一个输入安全组控制器,IPv4 CIDR块可以将内容推向推出类型的输入.

declare const stack: Stack;
const sg = new medialive.InputSecurityGroup(stack, 'SG', {
  allowlistRules: ['203.0.113.0/24'],
});

导入现有输入安全组

declare const stack: Stack;
const sg = medialive.InputSecurityGroup.fromInputSecurityGroupArn(stack, 'Imported',
  'arn:aws:medialive:us-east-1:123456789012:inputSecurityGroup:1234567');

频道

一个信道取出一个或多个输入,编码它们,并产生输出组. 若无 role 提供,频道自动创建 IAM 角色与 medialive.amazonaws.com 服务校长.

最小的例子——单输入,单HLS输出:

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;

const video = medialive.EncodeConfiguration.video({
  name: 'video_720p',
  codec: medialive.VideoCodecSettings.h264({
    rateControl: medialive.H264RateControl.cbr({ bitrate: Bitrate.mbps(3) }),
    framerate: medialive.Framerate.FPS_30,
  }),
  width: 1280,
  height: 720,
});

const audio = medialive.EncodeConfiguration.audio({
  name: 'audio_aac',
  codec: medialive.AudioCodecSettings.aac({ bitrate: Bitrate.kbps(192) }),
});

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
    }),
  ],
});

带有媒体包 V2 的 Standard 频道

一个标准达道运行两条管道进行冗余. 每个产出组需要两个目的地——每个管道一个.

declare const stack: Stack;
declare const input: medialive.IInput;
declare const mpChannel: mediapackagev2.IChannel;

const hdVideo = medialive.EncodeConfiguration.video({
  name: 'video_1080p',
  codec: medialive.VideoCodecSettings.h265({
    rateControl: medialive.H265RateControl.qvbr({
      maxBitrate: Bitrate.mbps(8),
      qvbrQualityLevel: 7,
    }),
    framerate: medialive.Framerate.FPS_30,
  }),
  width: 1920,
  height: 1080,
});

const sdVideo = medialive.EncodeConfiguration.video({
  name: 'video_480p',
  codec: medialive.VideoCodecSettings.h265({
    rateControl: medialive.H265RateControl.qvbr({
      maxBitrate: Bitrate.mbps(2),
      qvbrQualityLevel: 7,
    }),
    framerate: medialive.Framerate.FPS_30,
  }),
  width: 854,
  height: 480,
});

const audio = medialive.EncodeConfiguration.audio({
  name: 'audio_aac',
  codec: medialive.AudioCodecSettings.aac({ bitrate: Bitrate.kbps(192) }),
});

new medialive.Channel(stack, 'Channel', {
  channelClass: medialive.ChannelClass.STANDARD,
  inputs: [{ input }],
  outputGroups: [
    medialive.OutputGroupConfiguration.mediaPackageV2({
      name: 'emp',
      channel: mpChannel,
      outputs: [
        { encode: hdVideo, outputName: 'hd' },
        { encode: sdVideo, outputName: 'sd' },
        { encode: audio, outputName: 'audio' },
      ],
    }),
  ],
});

全局配置

globalConfiguration 设置全通道行为:管道如何被锁定在一起以及输出时间源。 所有字段都是可选的并倒回MediaLive默认.

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  timecodeConfig: {
    source: medialive.TimecodeSource.EMBEDDED,
  },
  globalConfiguration: {
    outputLocking: medialive.OutputLocking.epoch(),
    outputTimingSource: medialive.OutputTimingSource.INPUT_CLOCK,
  },
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
    }),
  ],
});

输出锁定

outputLocking 同步一个频道的两条管道发射的帧. 选择一个策略 页:1 OutputLocking 工厂:

• OutputLocking.pipeline() ——使每条管线的输出互相同步. 选择如何使用 method数字 : PipelineLockingMethod.SOURCE_TIMECODE (默认,需要可靠的嵌入式时间码)或 PipelineLockingMethod.VIDEO_ALIGNMENT (视频内容匹配,不需要时间码).
• OutputLocking.epoch() ——与 Unix 时代同步(可选 a) customEpoch编号 :jamSyncTime) (中文(简体) ). 要求数 outputTimingSource: OutputTimingSource.INPUT_CLOCK (在合成时强制).
• OutputLocking.disabled() ——不同步.

// Video-aligned pipeline locking — useful when sources lack reliable timecodes
const locking = medialive.OutputLocking.pipeline({
  method: medialive.PipelineLockingMethod.VIDEO_ALIGNMENT,
});

输入- 丢失行为

inputLossBehavior 当输入丢失时, 控制 MediaLive 释放出什么: 一个黑色的周期, 然后是一个 重复相框,然后是坚实的颜色或平板图像。 提供清单为 FileLocation。 。 。 .

declare const slateBucket: s3.IBucket;

const inputLoss: medialive.InputLossBehavior = {
  blackFrame: Duration.seconds(1),
  repeatFrame: Duration.seconds(5),
  imageType: medialive.InputLossImageType.SLATE,
  imageSlate: medialive.FileLocation.fromBucket(slateBucket, 'slates/offline.png'),
};

文件位置

多个频道的功能是参考一个文件MediaLive在运行时读取的——一个输入-丢失列表,一个 显示效果的图像、 断电的平板图像, 或刻录的字幕字体 。 这些都是一个 FileLocation,从 S3 桶(自动授予频道角色读取权限)或 URL 创建 (带有可选的SSM支持的证书):

import { StringParameter } from 'aws-cdk-lib/aws-ssm';

declare const bucket: s3.IBucket;
declare const passwordParam: StringParameter;

// From an S3 bucket — the channel role is granted read access automatically
const fromS3 = medialive.FileLocation.fromBucket(bucket, 'assets/slate.png');

// From a URL with optional credentials (SSM parameter read access auto-granted)
const fromUrl = medialive.FileLocation.url('https://origin.example.com/font.ttf', {
  username: 'ingest-user',
  password: passwordParam,
});

颜色校正

频道可以对其视频应用一个或多个色彩空间转换,可选使用3D LUT 重新绘制颜色。 每个 ColorCorrection 声明 inputColorSpace 来匹配和 outputColorSpace 转换为。 MediaLive在运行时从S3读取LUT,所以必须是S3 地点——通过 Lut.fromBucket() (使用安全设备) s3ssl:// 表格和自动赠款 频道角色读取访问)或 Lut.url() 带有 s3://编号 :s3ssl:// URL :

declare const stack: Stack;
declare const bucket: s3.IBucket;
declare const input: medialive.IInput;
declare const video: medialive.EncodeConfiguration;
declare const destination: medialive.OutputDestination;

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  colorCorrections: [{
    inputColorSpace: medialive.ColorSpace.REC_601,
    outputColorSpace: medialive.ColorSpace.REC_709,
    lut: medialive.Lut.fromBucket(bucket, 'luts/rec601-to-rec709.cube'),
  }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [destination],
      outputs: [{ encodes: [video], outputName: 'video' }],
    }),
  ],
});

编码配置

使用 EncodeConfiguration.video(), (中文(简体) ), EncodeConfiguration.audio(),以及 EncodeConfiguration.caption() 定义编码.

视频

// H.264
const h264 = medialive.EncodeConfiguration.video({
  name: 'h264_720p',
  codec: medialive.VideoCodecSettings.h264({
    rateControl: medialive.H264RateControl.cbr({ bitrate: Bitrate.mbps(3) }),
    framerate: medialive.Framerate.FPS_30,
    profile: medialive.H264Profile.HIGH,
  }),
  width: 1280,
  height: 720,
});

// H.265
const h265 = medialive.EncodeConfiguration.video({
  name: 'h265_1080p',
  codec: medialive.VideoCodecSettings.h265({
    rateControl: medialive.H265RateControl.qvbr({
      maxBitrate: Bitrate.mbps(5),
      qvbrQualityLevel: 7,
    }),
    framerate: medialive.Framerate.FPS_30,
    profile: medialive.H265Profile.MAIN,
    tier: medialive.H265Tier.HIGH,
  }),
  width: 1920,
  height: 1080,
});

视频解码器接受可选的覆盖,用于适应性量化,场景变化检测,色彩空间等. 参见完整列表的道具接口 :

const hdr = medialive.EncodeConfiguration.video({
  name: 'h265_hdr',
  codec: medialive.VideoCodecSettings.h265({
    rateControl: medialive.H265RateControl.qvbr({ maxBitrate: Bitrate.mbps(8), qvbrQualityLevel: 8 }),
    framerate: medialive.Framerate.FPS_30,
    sceneChangeDetect: medialive.H265SceneChangeDetect.ENABLED,
    colorSpaceSettings: medialive.H265ColorSpaceSettings.hlg2020(),
  }),
  width: 1920,
  height: 1080,
});

音频

// AAC stereo
const aac = medialive.EncodeConfiguration.audio({
  name: 'aac_stereo',
  codec: medialive.AudioCodecSettings.aac({
    bitrate: Bitrate.kbps(192),
    codingMode: medialive.AacCodingMode.CODING_MODE_2_0,
  }),
});

// AC3 5.1
const ac3 = medialive.EncodeConfiguration.audio({
  name: 'ac3_surround',
  codec: medialive.AudioCodecSettings.ac3({
    bitrate: Bitrate.kbps(384),
    codingMode: medialive.Ac3CodingMode.CODING_MODE_3_2_LFE,
  }),
});

说明

标题编码转换源标题音轨( 引用于 captionSelectorName改为: 输入格式 CaptionDestination 工厂。 一个选择器可以提供多个编码 :

// Define a caption selector on the input attachment (see Input Attachment Settings below)
const captionSelector = medialive.CaptionSelector.embedded('captions');

// WebVTT captions — packaged alongside the video encode in the same output
const webvtt = medialive.EncodeConfiguration.caption({
  name: 'eng_webvtt',
  captionSelectorName: captionSelector.name,
  languageCode: 'eng',
  languageDescription: 'English',
  destination: medialive.CaptionDestination.webvtt(),
});

// Burned-in captions — rendered into the video, styled via the burn-in options
const burnIn = medialive.EncodeConfiguration.caption({
  name: 'eng_burnin',
  captionSelectorName: captionSelector.name,
  destination: medialive.CaptionDestination.burnIn({
    alignment: medialive.CaptionAlignment.CENTERED,
    fontColor: medialive.CaptionFontColor.WHITE,
    outlineColor: medialive.CaptionOutlineColor.BLACK,
    fontSize: medialive.CaptionFontSize.AUTO,
  }),
});

跨服务整合

AWS 元素介质包 V2

使用 mediaPackageV2() 和通过一个单 channel —— MediaLive 将每个管道自动映射到一个 MediaPackage 摄入端点(一个用于 SINGLE_PIPELINE,两者都用于 STANDARD) (中文(简体) ). 每个输出包含一个单一的编码(每个输出一个音轨).

带内字幕(烧入、嵌入)与视频编码一起通过 captions 道具 :

declare const mpChannel: mediapackagev2.IChannel;
declare const hdVideo: medialive.EncodeConfiguration;
declare const sdVideo: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;
declare const burnIn: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.mediaPackageV2({
  name: 'emp',
  channel: mpChannel,
  outputs: [
    { encode: hdVideo, captions: [burnIn], outputName: 'hd' },
    { encode: sdVideo, outputName: 'sd' },
    { encode: audio, outputName: 'audio' },
  ],
});

每条管线的控制——例如将管线0钉入特定终点,或将每条管线输送到不同的(跨区域)通道——使用 mediaPackageV2PerPipeline() 明确目的地:

declare const primary: mediapackagev2.IChannel;
declare const hdVideo: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.mediaPackageV2PerPipeline({
  name: 'emp',
  destinations: [
    // destinations[0] → Pipeline 0, destinations[1] → Pipeline 1
    medialive.MediaPackageV2Destination.channel(primary, medialive.MediaPackageV2EndpointId.ENDPOINT_2),
    medialive.MediaPackageV2Destination.channel(primary, medialive.MediaPackageV2EndpointId.ENDPOINT_1),
  ],
  outputs: [
    { encode: hdVideo, outputName: 'hd' },
  ],
});

高级

使用 OutputDestination.url() hTTP 源代码或 OutputDestination.toBucket() 关于S3:

declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

// HLS to S3
medialive.OutputGroupConfiguration.hls({
  name: 'hls_s3',
  destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
  outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
});

// HLS to an HTTPS CDN origin.
medialive.OutputGroupConfiguration.hls({
  name: 'hls-http',
  destinations: [medialive.OutputDestination.url('https://203.0.113.10/ingest/stream')],
  hlsCdnSettings: medialive.HlsCdnSettings.basicPut(),
  outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
});

存档

归档输出将长式录音写入 S3 :

declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.archive({
  name: 'archive',
  destinations: [medialive.S3OutputDestination.toBucket(bucket, 'archive/recording')],
  rolloverInterval: Duration.seconds(600),
  outputs: [{ encodes: [video, audio], outputName: 'archive_out' }],
});

RTMP软件

RTMP的产出只支持H.264 + AAC. 每个输出每个信道管道(控制台的"Destination A"/"Destination B")需要1个目的地 RtmpDestination.url() - 一个给 SINGLE_PIPELINE,两个用于 STANDARD数字 :

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.rtmp({
  name: 'social',
  outputs: [{
    encodes: [video, audio],
    outputName: 'live',
    destinations: [
      medialive.RtmpDestination.url('rtmp://rtmp.example.com/live', 'your-stream-key'),
    ],
  }],
});

特别审判庭

SRT产出的使用 SrtDestination.caller() 用于呼叫模式或 SrtDestination.listener() 用于听众模式。 当您已经拥有完整的 SRT URL 而不是单独的主机和端口时,请使用 SrtDestination.callerUrl(). SRT 输出总是加密的,所以每个目的地都需要一个 encryptionPassphraseSecret (一个保密经理的秘密). 每种产出每个信道管道(“A”/“B”)需要一个目的地——一个用于 SINGLE_PIPELINE,两个用于 STANDARD数字 :

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;
declare const passphrase: secretsmanager.ISecret;

// SRT caller to a remote listener
medialive.OutputGroupConfiguration.srt({
  name: 'srt_out',
  outputs: [{
    encodes: [video, audio],
    outputName: 'srt_caller',
    destinations: [medialive.SrtDestination.caller({
      address: '203.0.113.20',
      port: 5000,
      encryptionPassphraseSecret: passphrase,
    })],
  }],
});

// SRT listener — MediaLive waits for the downstream system to connect
medialive.OutputGroupConfiguration.srt({
  name: 'srt_listen',
  outputs: [{
    encodes: [video, audio],
    outputName: 'srt_listener',
    destinations: [medialive.SrtDestination.listener({
      listenerPort: 5000,
      encryptionPassphraseSecret: passphrase,
    })],
  }],
});

AWS 元素介质连接路由器

mediaConnectRouter() 将每个信道管道发送到 AWS Elemental MediaConnect 路由器。 转运加密默认到AUTOMATIC;CDK从信道类中每个管线产生一个去向,因此常见的情况不需要每个管线配置. 您必须指定 availabilityZones - 一个是 SINGLE_PIPELINE 通道,或两个(每个管道一个)用于 STANDARD下游接线——每条输入输油管的路由器输入输油管的输油管——是在MediaConnect一侧配置的,以名称和输油管编号来参考这一组的产出.

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;
declare const passphrase: secretsmanager.ISecret;
declare const passphrase1: secretsmanager.ISecret;

// AUTOMATIC encryption on every pipeline (MPEG-TS container, like UDP)
medialive.OutputGroupConfiguration.mediaConnectRouter({
  name: 'router_out',
  availabilityZones: ['us-east-1a'],
  outputs: [{ encodes: [video, audio], outputName: 'router_ts' }],
});

// One shared Secrets Manager passphrase across all pipelines (SECRETS_MANAGER encryption)
medialive.OutputGroupConfiguration.mediaConnectRouter({
  name: 'router_out',
  availabilityZones: ['us-east-1a'],
  routerSettings: medialive.MediaConnectRouterSettings.shared({ encryptionSecret: passphrase }),
  outputs: [{ encodes: [video, audio], outputName: 'router_ts' }],
});

// Distinct encryption per pipeline — an omitted pipeline stays AUTOMATIC (STANDARD channels)
medialive.OutputGroupConfiguration.mediaConnectRouter({
  name: 'router_out',
  availabilityZones: ['us-east-1a', 'us-east-1b'],
  routerSettings: medialive.MediaConnectRouterSettings.perPipeline({
    pipeline1: { encryptionSecret: passphrase1 },
  }),
  outputs: [{ encodes: [video, audio], outputName: 'router_ts' }],
});

当提供密码口令密语时,该频道的IAM角色会自动获得读取权限.

UDP 服务器

UDP输出通过UDP或RTP提供MPEG-TS. 使用 UdpOutputDestination.udp() 用于纯 UDP 或 .rtp() 用于RTP(如果使用FEC需要):

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.udp({
  name: 'udp_out',
  destinations: [medialive.UdpOutputDestination.udp({ address: '203.0.113.5', port: 5000 })],
  outputs: [{ encodes: [video, audio], outputName: 'ts_out' }],
});

框架抓取

框架抓取输出将周期 JPEG 快照写入 S3 :

declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.frameCapture({
  name: 'thumbnails',
  destinations: [medialive.S3OutputDestination.toBucket(bucket, 'thumbnails/live')],
  outputs: [{ encodes: [video], outputName: 'thumb' }],
});

微软平滑流线

MS平滑输出将零碎的MP4推向IIS平滑流端:

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.msSmooth({
  name: 'smooth',
  destinations: [medialive.OutputDestination.url('https://smooth.example.com/live')],
  outputs: [{ encodes: [video, audio], outputName: 'smooth_out' }],
});

按输出的 HLS 设置

HLS 输出接受每个输出 hlsSettings 通过 HlsSettings 工厂 - standard() 视频 引渡(可选) M3u8Settings 用于运输, audioOnly() 音频 引渡(可选的涵盖条款作为 FileLocation, (中文(简体) ), fmp4(),或 frameCapture()。 。 。 .

declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.hls({
  name: 'hls',
  destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
  outputs: [
    {
      encodes: [video],
      outputName: 'video',
      hlsSettings: medialive.HlsSettings.standard({
        m3u8Settings: medialive.M3u8Settings.of({
          scte35Behavior: medialive.M3u8Scte35Behavior.PASSTHROUGH,
          programNum: 1,
        }),
      }),
    },
    {
      encodes: [audio],
      outputName: 'audio',
      hlsSettings: medialive.HlsSettings.audioOnly({
        audioGroupId: 'program',
        audioOnlyImage: medialive.FileLocation.fromBucket(bucket, 'art/cover.png'),
      }),
    },
  ],
});

转发错误校正( UDP)

UDP 输出接受可选 fec 设置(SMPTE 2022-1) – 只列或列并列 FEC. FEC 需要一个 rtp:// 目标 :

declare const video: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.udp({
  name: 'udp',
  destinations: [medialive.UdpOutputDestination.rtp({ address: '203.0.113.5', port: 5000 })],
  outputs: [{
    encodes: [video],
    outputName: 'ts',
    fec: { mode: medialive.FecMode.COLUMN_AND_ROW, columnDepth: 10, rowLength: 10 },
  }],
});

MPEG-TS 容器设置

MPEG-TS 输出组 - udp(), (中文(简体) ), archive(), (中文(简体) ), srt(),以及 mediaConnectRouter() ——接受可选产出 m2tsSettings 通过 M2tsSettings.of(). 启用它来使用 MediaLive 的服务默认值. 使用比特率 Bitrate,间隔使用 Duration,而闭合值字段使用enum(例如. M2tsRateMode, (中文(简体) ), M2tsScte35Control); PID 字段是接受十进制,十六进制,范围,或逗号分隔列表的字符串.

declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

medialive.OutputGroupConfiguration.udp({
  name: 'udp_out',
  destinations: [medialive.UdpOutputDestination.udp({ address: '203.0.113.5', port: 5000 })],
  outputs: [{
    encodes: [video, audio],
    outputName: 'ts',
    m2tsSettings: medialive.M2tsSettings.of({
      bitrate: Bitrate.mbps(8),
      rateMode: medialive.M2tsRateMode.VBR,
      programNum: 1,
      patInterval: Duration.millis(100),
      pmtInterval: Duration.millis(100),
      scte35Control: medialive.M2tsScte35Control.PASSTHROUGH,
      dvbSdtSettings: {
        outputSdt: medialive.DvbSdtOutputMode.SDT_MANUAL,
        serviceName: 'My Service',
        repInterval: Duration.millis(2000),
      },
    }),
  }],
});

目标

每个输出组类型都使用特定的目的地类. 通过静态工厂方法创建目的地:

OutputDestination.toBucket() (和 S3OutputDestination.toBucket()) 建造峡谷 s3ssl:// URL和自动授予频道的IAM角色所需的S3权限; InputSource.fromBucket() 输入读取时使用相同的输入. MediaPackageV2Destination.channel() 自动授予 MediaPackage V2 频道的摄入权限.

MediaConnect路由器输出组没有目的类——其发送被配置在MediaConnect一侧. 每条管道中转加密通过组设置 routerSettings 使用道具 MediaConnectRouterSettings.shared() 编号 : .perPipeline() (见 媒体连接路由器 (见上文).

其他目标

媒体包V2和CMAF 支持的有限产出组 additionalDestinations 用于跨区域交付或备份包装。 它们是与管道冗余分开的——它们把相同的内容传出额外的终点.

每个目的地的区域自动从频道的堆栈中解决. 就跨区域的进口而言,明确通过本区域:

declare const primaryChannel: mediapackagev2.IChannel;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

// Import a channel from another region — the region travels with the channel
const backupChannel = mediapackagev2.Channel.fromChannelAttributes(this, 'BackupChannel', {
  channelName: 'backup-channel',
  channelGroupName: 'backup-group',
  region: 'us-west-2',
});

medialive.OutputGroupConfiguration.mediaPackageV2({
  name: 'emp',
  channel: primaryChannel,
  additionalDestinations: [
    // Cross-region: the destination picks up us-west-2 from the imported channel
    medialive.MediaPackageV2Destination.channel(backupChannel, medialive.MediaPackageV2EndpointId.ENDPOINT_1),
  ],
  outputs: [
    { encode: video, outputName: 'video' },
    { encode: audio, outputName: 'audio' },
  ],
});

管道冗余

频道默认为 SINGLE_PIPELINE设置 channelClass: ChannelClass.STANDARD 双管冗余.

当使用标准:

• 每个输出组 destinations 数组必须有两个条目—— destinations[0] 地图到管道 0, destinations[1] 地图到管道 1.
• 对于MediaPackage V2, 使用 ENDPOINT_1 管道 0 和 ENDPOINT_2 1号管道.
• additionalDestinations 与管道冗余分开——它们向外延伸到额外的终点.

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;

new medialive.Channel(stack, 'StandardChannel', {
  channelClass: medialive.ChannelClass.STANDARD,
  inputs: [{ input }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [
        medialive.OutputDestination.toBucket(bucket, 'live/pipeline0'),
        medialive.OutputDestination.toBucket(bucket, 'live/pipeline1'),
      ],
      outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
    }),
  ],
});

输入附件设置

每个条目 inputs 是一个输入附件,它可以携带每输入提取和连接 输入本身以外的设置.

选择器 从输入中选择特定的音轨。 使用 AudioSelector (单位:千美元)byLanguage(), (中文(简体) ), byPid(), (中文(简体) ), byTrack(), (中文(简体) ), hlsRendition(), (中文(简体) ), default(), (中文(简体) ), CaptionSelector (单位:千美元)byLanguage(), (中文(简体) ), embedded(), (中文(简体) ), ancillary(), (中文(简体) ), dvbSub(), (中文(简体) ), scte27(), (中文(简体) ), teletext(), (中文(简体) ), arib()),以及 videoSelector (颜色空间, HDR10 元数据和通过 VideoSelection) (中文(简体) ). 标题编码后引用 a 标题选择器的名称 .

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;

new medialive.Channel(stack, 'Channel', {
  inputs: [{
    input,
    audioSelectors: [
      medialive.AudioSelector.byLanguage('eng', 'eng', medialive.AudioLanguageSelectionPolicy.STRICT),
    ],
    captionSelectors: [
      medialive.CaptionSelector.embedded('embedded'),
    ],
    videoSelector: {
      colorSpace: medialive.VideoColorSpace.HDR10,
      colorSpaceUsage: medialive.VideoColorSpaceUsage.FORCE,
      selectBy: medialive.VideoSelection.byProgramId(1),
    },
  }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video], outputName: 'hls_out' }],
    }),
  ],
});

网络输入设置 适用于 URL- pull 和多播输入 — HLS 带宽/缓冲/ 重试 行为,SCTE-35来源HlsScte35Source.SEGMENTS 或者说 MANIFEST),HTTPS服务器验证, 和用于源特定多播的多播源IP. logicalInterfaceNames 映射输入到 mediaLive Anywhere节点上的网络接口.

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;

new medialive.Channel(stack, 'Channel', {
  inputs: [{
    input,
    networkInputSettings: {
      serverValidation: medialive.ServerValidation.CHECK_CRYPTOGRAPHY_AND_VALIDATE_NAME,
      hlsInputSettings: {
        bandwidth: Bitrate.mbps(5),
        scte35Source: medialive.HlsScte35Source.MANIFEST,
      },
    },
    logicalInterfaceNames: ['eth0', 'eth1'],
  }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video], outputName: 'hls_out' }],
    }),
  ],
});

自动输入失败

自动输入失败给您输入 -来源 冗余: 附加二级输入,以及 MediaLive 当活动输入遇到故障时不重新启动频道切换到它 条件。 这与管道冗余无关 ChannelClass.STANDARD (哪个) 重复两个管道的单一来源.

提供 automaticInputFailover 输入附件。 如果你没有指定条件,a 使用单个输入- 丢失条件 :

declare const stack: Stack;
declare const primaryInput: medialive.IInput;
declare const secondaryInput: medialive.IInput;
declare const audioSelector: medialive.AudioSelector;
declare const video: medialive.EncodeConfiguration;
declare const audio: medialive.EncodeConfiguration;
declare const bucket: s3.IBucket;

new medialive.Channel(stack, 'Channel', {
  inputs: [{
    input: primaryInput,
    automaticInputFailover: {
      secondaryInput,
      inputPreference: medialive.InputPreference.PRIMARY_INPUT_PREFERRED,
      errorClearTime: Duration.seconds(3),
      failoverConditions: [
        medialive.FailoverCondition.inputLoss({ threshold: Duration.millis(1500) }),
        medialive.FailoverCondition.audioSilence({ audioSelector, threshold: Duration.seconds(2) }),
        medialive.FailoverCondition.videoBlack({ blackDetectThreshold: 0.1, threshold: Duration.seconds(1) }),
      ],
    },
  }, {
    // The secondary input must also be attached to the channel as its own input.
    input: secondaryInput,
  }],
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video, audio], outputName: 'hls_out' }],
    }),
  ],
});

初级和二级投入必须具有相同的投入类别。 该频道的IAM角色是 自动获得读取二次输入源,就像主输入一样.

Ad Avail 处理

MediaLive在广告使用时可以空出内容,插入停电板并信号SCTE-35广告使用 到下游系统。 这些都是频道级道具.

availBlanking 以黑色(或图像)取代广告使用时的视频/音频/封装 blackoutSlate 显示 SCTE- 35 断电时的列表 。 两个图像字段都使用 FileLocation。 。 。 .

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  availBlanking: {
    state: medialive.AvailBlankingState.ENABLED,
    image: medialive.FileLocation.fromBucket(bucket, 'slates/avail.png'),
  },
  blackoutSlate: {
    state: medialive.BlackoutSlateState.ENABLED,
    image: medialive.FileLocation.fromBucket(bucket, 'slates/blackout.png'),
  },
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video], outputName: 'hls_out' }],
    }),
  ],
});

availSettings 选择如何处理 SCTE- 35 广告效果 - AvailSettings.spliceInsert(), (中文(简体) ), AvailSettings.timeSignalApos(),或 AvailSettings.esam() 用于事件信号和管理 与外部 POIS 端点相对. scte35SegmentationScope 输出组收到的控件 分道扬镳 ESAM POIS 密码作为 SSM 参数提供,频道角色 可以自动读取.

import { StringParameter } from 'aws-cdk-lib/aws-ssm';

declare const stack: Stack;
declare const input: medialive.IInput;
declare const bucket: s3.IBucket;
declare const video: medialive.EncodeConfiguration;
declare const poisPassword: StringParameter;

new medialive.Channel(stack, 'Channel', {
  inputs: [{ input }],
  availSettings: medialive.AvailSettings.esam({
    pois: {
      url: 'https://pois.example.com/esam',
      username: 'pois-user',
      password: poisPassword,
    },
    acquisitionPointId: 'acquisition-point-1',
    adAvailOffset: Duration.millis(200),
  }),
  scte35SegmentationScope: medialive.Scte35SegmentationScope.SCTE35_ENABLED_OUTPUT_GROUPS,
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video], outputName: 'hls_out' }],
    }),
  ],
});

自动创建的角色和赠款

否时 role 提供,频道自动创建 IAM 角色与 medialive.amazonaws.com 服务本金并只授予您实际需要的配置权限。 这些自动赠款适用 仅限 到频道管理的角色; 如果您带来了自己的 role,没有添加.

频道角色补助金 ——基于您配置的线(仅由频道管理的角色) :

投入角色赠款 ——与频道角色相分离,用于输入创建/删除时间. 和频道角色一样,这些只有在输入自动创建其角色时才会被添加; 通过 a role 改为 mediaConnect() 或者说 cdi() 并且不增加补助金:

渠道和输入自动创建的角色都包括相互混淆的预防(aws:SourceAccount + 键 aws:SourceArn (第6条)。 关于信任实体要求的全部清单,见 文件。 。 。 .

自动创建的角色可以在 channel.role 如果您需要添加更多权限.

带来你预定义的角色

当你通过一个 role,频道制作 没有 自动赠款——您需要添加角色所需的权限. 这既包括主要政策,也包括任何参考资源政策:S3输出目的地和输入源、秘密管理器和SSM读取、MediaPackage V2摄取、Cloud Watch Logs和VPC输出ENI管理。 见 信托实体要求,或通过账户 MediaLiveAccessRole ——MediaLive可以承担的IAM角色.

云表计量

频道显示在 AWS/MediaLive 名称空间, 以 ChannelId 财务报告和审定财务报表 Pipeline。将下面的指定助手用于最常用的衡量标准,或 metric(metricName, pipeline) ,以获取 媒体计量参考。 。 。 .

MediaLive每条管道公布衡量标准。 每个帮手都拿一个 Pipeline 所以你要明确决定 你所监视的输油管. STANDARD 频道运行两条冗余管道(PIPELINE_0, (中文(简体) ), PIPELINE_1) ——同时发出警报以覆盖整个频道. SINGLE_PIPELINE 频道只发布于 PIPELINE_0; 通过 PIPELINE_1 扔在合成时间.

declare const channel: medialive.Channel;
declare const stack: Stack;

channel.metricDroppedFrames(medialive.Pipeline.PIPELINE_0).createAlarm(stack, 'DroppedFrames', {
  threshold: 1,
  evaluationPeriods: 2,
});

channel.metricSvqTime(medialive.Pipeline.PIPELINE_0).createAlarm(stack, 'SvqTime', {
  threshold: 0,
  evaluationPeriods: 1,
});

// Custom metric by name with sum statistic
channel.metric('Output4xxErrors', medialive.Pipeline.PIPELINE_0, { statistic: 'sum' });

对于标准达道,警报 在两个管道:

declare const standardChannel: medialive.Channel;
declare const stack: Stack;

standardChannel.metricDroppedFrames(medialive.Pipeline.PIPELINE_0).createAlarm(stack, 'Drops0', {
  threshold: 1,
  evaluationPeriods: 2,
});
standardChannel.metricDroppedFrames(medialive.Pipeline.PIPELINE_1).createAlarm(stack, 'Drops1', {
  threshold: 1,
  evaluationPeriods: 2,
});

频道指标

默认值与每个度量的 AWS 推荐统计值相匹配 。 过 props 覆盖统计、周期、维度或任何其他 MetricOptions 字段.

媒体在任何地方

媒体生活 任何地方都允许你运行 MediaLive 频道 在自己的前提硬件.

某些输入类型仅可用任意频道(频道配置为 anywhereSettings: SDI,SMPTE 2110接收器组,和多播. 试图在云道上使用这些输入类型,会在合成时抛出一个验证错误.

网络

一个网络定义了任何地方资源的IP地址池和路径:

declare const stack: Stack;
const network = new medialive.Network(stack, 'Network', {
  networkName: 'on-prem-network',
  ipPools: ['10.0.0.0/24'],
  routes: [{ cidr: '0.0.0.0/0', gateway: '10.0.0.1' }],
});

组

一组代表一组关于前提的硬件节点:

declare const stack: Stack;
declare const instanceRole: iam.IRole;

const cluster = new medialive.Cluster(stack, 'Cluster', {
  clusterName: 'on-prem-cluster',
  clusterType: medialive.ClusterType.ON_PREMISES,
  instanceRole,
});

频道定位组

一个频道放置组将频道分配到一个集群内的特定节点. 把它与一个频道通过 anywhereSettings数字 :

declare const stack: Stack;
declare const cluster: medialive.ICluster;
declare const input: medialive.IInput;
declare const video: medialive.EncodeConfiguration;
declare const bucket: s3.IBucket;

const cpg = new medialive.ChannelPlacementGroup(stack, 'CPG', {
  channelPlacementGroupName: 'my-cpg',
  cluster,
});

new medialive.Channel(stack, 'AnywhereChannel', {
  inputs: [{ input }],
  anywhereSettings: { cluster, channelPlacementGroup: cpg },
  outputGroups: [
    medialive.OutputGroupConfiguration.hls({
      name: 'hls',
      destinations: [medialive.OutputDestination.toBucket(bucket, 'live/stream')],
      outputs: [{ encodes: [video], outputName: 'hls_out' }],
    }),
  ],
});

SDI 来源

一个 SDI 源代表任何硬件上的物理 SDI 输入 :

declare const stack: Stack;
const sdi = new medialive.SdiSource(stack, 'Sdi', {
  sdiSourceName: 'camera-1',
  type: medialive.SdiType.SINGLE,
});

前提输入网络

对于生活在网络上的投入,设置 inputNetworkLocation 改为 InputNetworkLocation.ON_PREMISES中输入不使用输入安全组。 推开 输入(RTMP/RTP/UDP)可将其目的地定在 Network,宣布 networkRoutes 改为 访问本地网络,并请求 staticIpAddress数字 :

declare const stack: Stack;

const network = new medialive.Network(stack, 'Network', {
  networkName: 'on-prem-network',
  ipPools: ['192.168.1.0/24'],
});

new medialive.Input(stack, 'OnPremInput', {
  inputName: 'on-prem-rtp',
  inputNetworkLocation: medialive.InputNetworkLocation.ON_PREMISES,
  input: medialive.InputConfiguration.rtpPush({
    destinations: [{
      network,
      networkRoutes: [{ cidr: '10.0.0.0/24', gateway: '10.0.0.1' }],
      staticIpAddress: '192.168.1.50',
    }],
  }),
});

SRT 听器输入接受 a streamId 上游系统在连接时使用:

declare const stack: Stack;
declare const sg: medialive.IInputSecurityGroup;

new medialive.Input(stack, 'SrtListener', {
  inputName: 'srt-listener',
  input: medialive.InputConfiguration.srtListener({
    inputSecurityGroups: [sg],
    streamId: 'my-stream-id',
  }),
});