如何在AWS MediaConvert中配置无音频输出并解决选择器报错

错误原因 #

Invalid selector_sequence_id [0] specified for audio_description [1]

你的输入是图片（无任何音频轨道），但配置里却定义了音频选择器和音频描述，MediaConvert找不到对应的音频资源，因此抛出该错误。

修改后的完整配置 #

{
    "middlewareStack": {},
    "input": {
        "Queue": "arn:aws:mediaconvert:eu-central-1:634617701827:queues/Default",
        "UserMetadata": {},
        "Role": "arn:aws:iam::634617701827:role/service-role/MediaConvert_Default_Role",
        "Settings": {
            "TimecodeConfig": {
                "Anchor": "00:00:00:00",
                "Source": "ZEROBASED"
            },
            "OutputGroups": [
                {
                    "Name": "File Group",
                    "Outputs": [
                        {
                            "Preset": "createPromoVideo",
                            "Extension": "mp4",
                            "NameModifier": "_fade",
                            "VideoDescription": {
                                "ScalingBehavior": "DEFAULT",
                                "TimecodeInsertion": "DISABLED",
                                "AntiAlias": "ENABLED",
                                "Sharpness": 50,
                                "Height": 1080,
                                "Width": 1080
                            }
                        }
                    ],
                    "OutputGroupSettings": {
                        "Type": "FILE_GROUP_SETTINGS",
                        "FileGroupSettings": {
                            "Destination": "s3://t44-post-cover/8fui.mp4",
                            "DestinationSettings": {
                                "S3Settings": {
                                    "AccessControl": {
                                        "CannedAcl": "PUBLIC_READ"
                                    }
                                }
                            }
                        }
                    }
                }
            ],
            "Inputs": [
                {
                    "FileInput": "s3://t44-post-cover/8fui",
                    "VideoSelector": {
                        "ColorSpace": "FOLLOW"
                    },
                    "FilterEnable": "AUTO",
                    "TimecodeSource": "ZEROBASED",
                    "InputClippings": [
                        {
                            "StartTimecode": "00:00:00:00",
                            "EndTimecode": "00:00:02:00"
                        }
                    ],
                    "FilterGraph": "fade=in:0:30"
                }
            ]
        },
        "AccelerationSettings": {
            "Mode": "DISABLED"
        },
        "StatusUpdateInterval": "SECONDS_60",
        "Priority": 0
    }
}

关键修改点 #

• 移除所有音频相关配置：删掉了输出端的AudioDescriptions和输入端的AudioSelectors，彻底取消音频处理流程
• 修正时间码源：将全局时间码源从EMBEDDED改为ZEROBASED，适配图片无嵌入时间码的特性
• 移除多余滤镜：删掉了视频描述里的淡出滤镜，只保留输入端的淡入效果，符合你的需求
• 保留核心配置：保留了图片2秒循环的裁剪设置、分辨率参数、S3输出路径等核心配置

内容的提问来源于stack exchange，提问作者János