AWS for M&E Blog
How to modify an incoming audio source using AWS Elemental MediaLive
Audio is a crucial component in media, elevating content beyond the visual and evoking a wide range of emotions. However, misconfiguring audio can negatively impact how your audience perceives and comprehends your content, resulting in a subpar user experience. Therefore, it’s essential to configure audio correctly in media to ensure your content is fully appreciated.
There may be instances where the incoming audio source doesn’t meet your output requirements or situations where you want to be more creative with your audio output. In such cases, AWS Elemental MediaLive, a broadcast-grade live video transcoding service, offers a solution through its audio channel remix setting.
The audio channel remix setting provides several ways to modify incoming audio signal to support various use cases:
- Reordering Channels: This allows you to change the order of channels within an audio track. A channel represents each audio source within the audio signal. For instance, a single microphone can produce one channel of audio, and an audio track combines multiple channels within a single audio signal, such as a stereo audio signal with Left and Right channels.
- Routing Channels: You can move audio channels from one or more input tracks to different output tracks.
- Combining Channels: audio from multiple channels can be combined into a single channel.
- Splitting Channels: the audio from a single channel can be splited into multiple channels.
- Adjusting the loudness: the loudness level of audio channels can be adjusted.
Understanding MediaLive audio configuration terminology
The following guide describes the use of this setting within AWS Elemental MediaLive.
Before you begin, it’s important to understand terminology in the MediaLive audio configuration. There are various categories to consider:
Input setup
Audio Selectors: These are sections within the audio input configuration where you define and encapsulate your audio input settings for easy identification and selection at the audio output side. Within audio selectors, you’ll find:
Audio Selector Name: A unique identifier for your input.
Selector Settings: Options for configuring your audio input format, including Audio HLS rendition, Audio Language, Audio PID, and Track.
Audio HLS Rendition: Specify a specific audio rendition within the HLS manifest by providing the GROUP-ID and NAME from the #EXT-X-MEDIA tag of the target HLS audio rendition (applicable for HLS input type only).
Audio Language: Choose a specific audio track based on an audio descriptor tagged with a language using a three-letter language code (applicable for MPEG transport streams input type only).
Audio PID: Select a specific audio track tagged with a Packet Identifier (PID) value within the source (applicable for MPEG transport streams input type only).
Audio Track: Use a specific audio track within the source (applicable for all input types).
Output setup
Remix Settings: This section allows you to configure your audio output settings, including:
Channels In: Represents the number of audio channels in the input (ranging from 1 to 16).
Channels Out: Specifies the number of output channels to be produced.
Channel Mapping: Allows you to map inputs to output tracks and adjust the gain or volume for the output channels. Channel mapping includes:
Output Channel: The index of the output channel being produced.
Input Channel Level: The index of the input from the source and the gain adjustment for a specific output channel.
Input Channel: The index of the input channel used as a source.
Gain: A value in decibels (from -60dB for mute to +6dB) to control audio gain.
Getting started
Let’s walk through how to use the audio remix functionality with a practical example. Suppose you have a source where the center speaker’s audio is quieter than desired, reducing voice clarity during action scenes. Additionally, due to a post-production error, the channel layout order in the 5.1 surround audio signal has been altered, with the center and LFE channels swapping positions.
The goal of this exercise is to address these issues.
Disclaimer: This guide assumes you have a basic understanding of creating a channel in MediaLive. If you’re unfamiliar with this topic, please review the guide here: [Link to Creating a Channel in MediaLive Guide].
Access your MediaLive channel and navigate to Input Attachments, selecting the input where you want to make changes.
In the General Input Settings section, under Audio Selector, click “Add Audio Selector.”
A new window will appear, allowing you to name this specific audio selector configuration and select the audio configuration type for input audio selection. For this case, we’ll use “Audio PID Selection.”
To proceed, it’s essential to understand your input audio configuration. If you’re unsure, you can use tools like VLC to inspect your audio characteristics. Inspecting the sample content in this case, the audio is defined as follows:
The interesting parts are:
ID or PID: This is the ID that allows you to select this audio in MediaLive.
Codec: Identifies the audio codec.
Channels: Indicates the number of channels and layout within the audio elementary stream.
For this example, the PID is 4352, the audio codec is Dolby E-AC3, and there are 3/2/LFE channels in a 5.1 format (L, R, C, LFE, Ls, Rs).
Returning to MediaLive’s configuration, navigate to your Audio Selector at the input side, give it a name, and select PID. In this case, the PID value is 4352.
Next, go to your Output Group => Stream Settings and select audio 1. Here, you’ll define the audio configuration for your audio stream. In the dropdown, choose the Audio Selector name from the input and the codec you want to use.
Now, use the remix setting. As mentioned at the beginning of this guide, the goal in this case is to fix the channel layout order between the Center and LFE channels and adjust the loudness gain in all channels. Turn on the remix setting, select the number of channels for your input and output (in this example, it’s a 5.1 layout, so 6 channels), and add the number of channels you’re mapping (also 6).
Changing the channel layout order is straightforward. In a typical 5.1 layout (L, R, C, LFE, Ls, Rs), the issue is that the Center (C) and LFE are in the wrong positions. C is currently in position 4, and LFE is in position 3. To correct this, in the Channel Mapping settings, specify that the output channel to be modified is the 3rd, but the input is currently in the 4th position. To correct LFE, the output location is the 4th, but the input is in the 3rd position. This action will produce an output track where the positions are swapped. Be sure to complete the entire input-to-output matrix for this 5.1 layout and add the correct input channel levels.
For the final part of this exercise, correct the audio gain for the center channel by adding 3 decibels. This will increase the dialogue volume, preventing it from being drowned out by noisy scenes. Map the audio channel output to the input that needs correction and set the gain to 3 dB. This setting allows you to modify the gain within the range of -60 dB (mute) to +6 dB. The example is for the center channel, typically located in the third position of the 5.1 surround layout. Be sure to complete the entire input-to-output matrix 5.1 layout but leaving the Gain blank to avoid making any changes.
Finally, click the “Create” button, and if everything is configured correctly, your channel is ready to start.
Conclusion
Now that you’ve seen how the remix setting works, you can explore this functionality in your own environment, applying it to various use cases. If you’d like to learn more about MediaLive-supported audio codecs for inputs and outputs, visit the following links: