Dolby® Encoding and Monitoring
Last updated
Last updated
The LA-5300 provides standard Dolby AC-4 encoding in support of ATSC 3.0 for its primary program (Program 1).
Optionally, it can encode a second program (Program 2) to Dolby Digital Plus (E-AC-3) and Dolby Digital (AC-3) to provide support for ATSC 1.0 services.
Clicking on the Dolby AC-4 Encoder dropdown menu from the Program 1 screen reveals the basic encoder controls. There is a separate dropdown menu for advanced controls.
Clicking on Program 2 in units so equipped provides similar controls for the Dolby Digital Plus encoder.
Basic parameters include Channel Mode (1A), Frame Rate (1B), Bit Rate (1C), audio Type (1G), Loudness Regulation (1F), Language (1E), and Dialogue Normalization (1D).
The Channel Mode control (1A) sets the configuration of the audio output channels embedded in the AC-4 bitstream. Choices include 2-channel (stereo), 5.1-channel, or 5.1.4-channel outputs (5.1-channel “base layer” with ear-level channels + 4 overhead channels).
The Dolby AC-4-encoded audio frames must be aligned with video frames to prevent losing audio frames or introducing A/V sync errors during source switching, much like Dolby E or ED2. The frame rate control (1B) should be set to match the video frame rate in your particular workflow. Select “Native” when there is no video reference available.
The Dolby AC-4 codec provides increased efficiency compared with Dolby AC-3 (Dolby Digital) to allow for the delivery of Dolby Atmos and multi-language content. The optimal bitrate for any given application – set by the Bit Rate control (1C) - will depend on a number of variables including channel mode and additional dialogue and music and effects content in NGA applications, but the basic recommendations are listed below.
AC-4 can deliver a traditional single mix, consisting of dialogue, music, and effects elements. This is referred to as a “complete mix” and often abbreviated as “CM”.
It can also deliver a single common music and effects element (“M&E”) together with separate dialogue elements (“D”) with each dialogue element selected and mixed with the M&E element in the AC-4 decoder. This ability is what allows for “personalized audio” as different elements can be combined into multiple audio presentations and delivered within a single bitstream.
Local emergency audio can also be delivered in the bitstream.
The Type control (1G) identifies whether the stream is Complete Main, Music and Effects, or Dialog. Currently, only Complete Main is supported.
Note - Emergency Audio is signaled by selecting EAS audio as the Priority 1 source when configuring input groups. Please see the section on Routing Audio to the Program Inputs for more information.
The AC-4 codec includes integrated loudness management via the Dolby Real Time Loudness Leveler. RTLL normalizes the incoming audio signal prior to encoding. Presets provided in the Loudness Practice control (1F) for the following regulations:
EBU R 128: Used primarily in Europe, audio is normalized to a target of -23dB LUFS as calculated over the entire duration of a program (integrated loudness) and without regard for isolating and measuring dialogue; it specifies a deviation of +/- 0.5LU and a peak level below -1dBTP
ATSC A/85: Used in the United States and serving as the foundation and reference for the CALM Act; whenever possible, the anchor element of the audio (typically dialogue) should be measured and normalized to a target of -24dB LKFS, +/- 2dB which is accomplished here by enabling the dialogue intelligence feature
ARIB TR-B32: The Japanese broadcast standard that uses a relative gate (per ITU BS 1770-2, like EBU R 128) but with a target loudness level of -24 LKFS and a maximum True Peak of -1dB.
Free TV OP-59: Used in Australia, this standard is based on the previous OP-48 regulations (which measured VU and digital peak levels and required a target level of -20dBFS) but instead measures average perceived loudness using ITU-R BS 1770-3 with a target of -24dB LKFS and a maximum True Peak of -2dB.
Setting the control to “Manual” allows the target loudness value to be manually set with the Dialogue Normalization control.
When using one of the pre-defined Loudness Practice profiles, the Dialogue Normalization control (1D) is set to “Auto” and grayed out, and dialogue normalization will be enabled or disabled per the selected profile. When the Loudness Practice is set to “Manual”, a specific loudness target can be set.
In addition, the Dialogue Intelligence control – described below in the “Advanced” menu section – can be turned on or off.
While the AC-4 codec supports the delivery of content in multiple languages, the primary language must be identified using the language dropdown menu (1E).
The advanced parameters menu contains controls that determine whether certain settings and metadata values are set automatically or manually, and, if set manually, the values of individual settings.
Please note the advanced menu scrolls within the web page and not all controls are visible at once. The screenshots and control descriptions below are divided into “top” and “bottom” figures for ease of illustration and explanation.
The Preferred Downmix Method control (2A) sets the downmixing metadata value for the downstream decoder.
Auto: Uses the upstream metadata to set the preferred downmix
Not Indicated: No instructions to the downstream decoder are specified, allowing the decoder or the end user to choose the downmix type
Lo/Ro: The Center and surround channel content is redirected to the Left and Right channels for playback in a system with only two speakers
PL: This is a Dolby Surround-compatible matrix-encoded downmix that contains content from all channels and downmixes them to the Left and Right channels; this downmix is intended to be decoded by a Dolby Surround Pro Logic decoder which can extract the matrix-encoded content from the Left and Right channels to produce a four- or five-channel output with mono surround channels
PLII: This is a Lt/Rt downmix that retains the stereo surround information from the original program; a Dolby Pro Logic II-compatible downmix is intended to be decoded by a Dolby Pro Logic II decoder.
The Settings Mode control (2B) determines whether the device-specific DRC profiles (1C) and various channel-specific downmix values (2D) are automatically or manually set.
When set to Auto, controls for device-specific DRC profiles as well as downmix modes are grayed out.
Selecting “Manual” allows DRC profiles to be individually set for Flat Panel televisions, Home Theater setups, Portable Headphones, and Portable Speakers (2C).
DRC profiles include:
Auto
None
Film Standard
Film Light
Music Standard
Music Light
Speech
Likewise, downmix gain values ranging from -6dB to +3dB can be individually specified Lo/Ro Surround, Lo/Ro Center, Lt/Rt Surround, and Lt/Rt Center Downmix configurations from a range of choices in each dropdown menu. Setting any of these controls to “Auto” will allow them to be set by upstream metadata.
Setting the Preprocessing Mode control (3A) to “Auto” grays out the associated controls in the “Previous Parameters” (3B) section and uses upstream metadata to determine individual control settings. Placing it in “Manual” allows each parameter to be manually and individually controlled.
“Applied” indicates LFE filtering has already been applied upstream and will not be added here. “Not Applied” indicates the filtering has not been applied upstream and will be added by the encoder. “Unknown” should be selected in situations where it is not known whether or not filtering has been applied.
“Applied” indicates a 90-degree phase shift has already been applied to the surround channels upstream and will not be added here. “Not Applied” indicates the phase shift has not been applied upstream and will be added by the encoder. “Unknown” should be selected in situations where it is not known whether or not the phase shift has been applied.
“Applied” indicates surround attenuation has already been applied upstream and will not be added here. “Not Applied” indicates surround attenuation has not been applied upstream and will be added by the encoder. “Unknown” should be selected in situations where it is not known whether or not surround attenuation has been applied.
If the downmix type is known and present in the incoming metadata stream, set this control to “Auto” to pass the metadata through to the decoder. Choose “Unknown” in situations where the previous mix type is not known.
Choosing “Mix Down Lo/Ro”, “Mix Down PL”, or “Mix Down PLII” will force the decoder to use the specified mix type.
5-channel content may have been initially downmixed from a higher channel count or upmixed from a lower channel count. If this information is known and present in the incoming metadata stream, set this control to “Auto” to pass the metadata through to the decoder. Choose “Unknown” if the previous mix type is not known.
Selecting “Mix Down PLIIx”, “Mix Down PLIIx Movie”, “Mix Down PLIIx Music”, or “Mix Down PLIIz” will force the decoder to use the specified mix type for content that has been mixed down from a high channel count.
Choosing “Mix Up PL”, “Mix Up PLII Movie”, “Mix Up PLII Music”, or “Mix Up PLII Professional” will force the decoder to use the specified mix type for content that has been upmixed from a lower channel count.
LFE signals are typically boosted by 10dB during the encoding process to ensure effects in the LFE channel are properly presented in the mix and deliver the desired impact. If this boost has already been applied upstream, select “Original Level” in the LFE Monitor Level control (3F). If it has not been applied upstream, select “Boost +10dB”. Selecting “Not Indicated” provides no specific information on this parameter to the downstream decoder.
The setting of the Dialogue Intelligence control (3E) is automatically set to comply with the specific loudness regulation selected with the Loudness Practice control. However, it can always be manually enabled and disabled.
Note that as soon as a different Loudness Practice is selected, the Dialogue Intelligence control will reflect the settings the chosen profile.
When using one of the included Loudness Practice profiles, the amount of loudness control provided by the Dolby Real Time Loudness Leveler is automatically set to deliver a compliant output and the Loudness Control Amount control (3C) will be grayed out.
When the Loudness Practice control is set to “Manual”, the amount of processing performed by the RTLL can be manually set. Lower settings provide more gentle level control and are best suited for content that has already been processed for compliance either in the file domain or by a real-time processor upstream. Higher settings gradually increase the amount of processing and are more suitable when program content levels vary. A setting of “0” defeats the RTLL completely.
Peak loudness values can be read on a sample-by-sample basis or with the use of over-sampling for a more granular measurement. Setting the Loudness Limit Mode control (3D) to “True Peak” will over-sample peak measurements in compliance with ITU-R BS.1770-4 recommendations.
The LA-5300 includes a Dolby AC-4 decoder for confidence monitoring. Monitor audio is routed to the channels immediately following those used for the bitstream output. For example, if the AC-4 bitstream occupies output channels 1 and 2 and the confidence decoder’s channel mode is 5.1, the monitor audio would be present on output channels 3 through 8.
Bear in mind that the controls in the Monitor menu only affect the monitor outputs and not actual bitstream output delivered to downstream facilities or viewers.
The Channel Mode control (4B) sets the channel mode, including 5.1- and 5.1.2-channel surround formats and stereo downmixes in Lo/Ro, Lt/Rt, PLII Lt/Rt, or Headphone.
The Target Reference Level control (4A) is used to emulate how consumer-side decoders scale the audio. For example, -31dBFS represents Line Mode, whereas -16 or -18dBFS would be useful for mobile-optimized streams.
Setting the Enable DRC control (4C) to “On” applies the DRC calculated by the encoder to the decoded signal.
To aid in dialogue intelligibility, the AC-4 codec allows signals that are either flagged with metadata as dialogue or determined to be dialogue by signal analysis to be boosted by the consumer. The Dialogue Enhancement Gain control (4E) sets the amount by which dialogue is boosted (or cut).
When dialogue gain is boosted or cut with the Dialogue Enhancement Gain control, the overall loudness will change compared to the desired target output level. Enabling the Dialogue Enhancement Preserve Loudness control (4D) will compensate for this and ensure the overall level matches the loudness target.
Clicking on the Dolby Digital Plus Encoder dropdown menu from the Program 2 screen reveals the basic encoder controls. There is a separate dropdown menu for advanced controls.
Basic parameters include Encoder Mode (5A), Channel Mode (5B), Bit Rate (5D), Loudness Method (5E), and Dialogue Normalization (5C).
Use the Encode Mode control (5A) to choose between Dolby Digital Plus (E-AC-3) or Dolby Digital (AC-3). For ATSC 1.0 applications, Dolby Digital is the proper choice.
The Channel Mode control (5B) determines the channel output mode of the encoder, either 2.0 (L,R) for 2-channel content, or 5.1 (L,R,C,LFE,Ls,Rs) for 5.1-channel content.
The optimal bitrate for any given application – set by the Bit Rate control (5D) - will depend on a number of variables including channel mode, available bandwidth, and desired audio quality. The default bitrates are listed below.
The Loudness Method control (5E) determines which processing algorithm is used to achieve an ATSC A/85-compliant output.
Linear Acoustic APTO is the same algorithm used in the Linear Acoustic ARC television processor which provides dynamic range processing and loudness leveling to transparently deliver audio that is both appropriately dynamic (to maintain the creative intent of the program producers) and well-controlled (to ensure CALM-compliance).
The Dolby Real Time Loudness Leveler, the same algorithm used in the Dolby AC-4 codec, normalizes the incoming audio signal to achieve ATSC A/85 compliance.
The Dialogue Normalization control (5C) sets the dialnorm metadata value in the Dolby AC-3 or E-AC-3 bitstream, which is the same as the desired output loudness target level. For ATSC A/85, this is normally set to -24dB.
The advanced parameters menu contains controls that determine whether certain settings and metadata values are set automatically or manually, and, if set manually, the values of individual settings.
The Preferred Downmix Method control (6A) sets the downmixing metadata value for the downstream decoder.
Not Indicated: No instructions to the downstream decoder are specified, allowing the decoder or the end user to choose the downmix type
Lt/Rt: The surround channels are added in-phase to the Left channel and out-of-phase to the Right channel, allowing a Dolby Surround Pro Logic decoder to reconstruct the Left, Center, Right, and Surround channels
Lo/Ro: The Center and surround channel content is redirected to the Left and Right channels for playback in a system with only two speakers
The Line Mode control (6B) sets the DRC (dynamic range control) profile for the line level output on home decoders and set top boxes. It typically uses a profile that uses lighter compression and allows the viewer some latitude in adjusting and scaling the amount of dynamic range as appropriate for their individual listening environment.
The RF Mode control (6C) sets the DRC profile for the RF input on a television, usually through the antenna output of a set top box. It typically uses a profile that employs a greater degree of compression and limiting compared to Line Mode, as the RF Mode profile is often used for the “midnight mode” on decoders.
Profiles for both the Line Mode and RF Mode include:
Film Standard: Provides a 5dB null band around the loudness target and offers consistent loudness for most television content while preserving some dynamics
Film Light: Provides a 20dB null band around the loudness target and delivers a much more dynamic and theatrical presentation than Film Standard
Music Standard: Provides a 5dB null band around the loudness target and uses higher ratios than the Film profiles to more effectively manage highly produced music formats
Music Light: Provides a 12dB null band around the loudness target and uses lower ratios to preserve the dynamics of genres such as classical and jazz
Speech: Provides a 5dB null band but higher ratios to more effectively deal with the varying levels and high peaks typically found in speech
The Downmix Gain controls (6D) can be individually set for the Lo/Ro Surround, Lo/Ro Center, Lt/Rt Surround, and Lt/Rt Center channels.
Values for the Surround channels range from -1.5dB through -6.0dB and –Inf, which disables the Surround channels. Values for the Center channels range from -6.0dB through +3.0db.
The LFE Filter control (6E) applies low-pass filter (LPF) at 250Hz in the LFE channel.
The Phase 90 Filter control (6F) applies a 90-degree phase shift to the Surround channels, allowing the downstream decoder to create a Lt/Rt downmix which can be rendered as an L, C, R, S signal by a Dolby Pro Logic encoder.
Enabling the Surround Attenuation control (6G) applies 3dB of attenuation to the Surround channels. This is primarily included for use in cinema applications where multiple surround speakers sum acoustically and can become louder than intended. The recommended setting for broadcast applications is “Disabled.”
Channel
Configuration
Recommended
Dolby AC-4 Data Rate
Equivalent
Dolby AC-3 Data Rate
Stereo (2/0)
64 kbps
192 kbps
5.1
144 kbps
382 kbps
5.1.4
288 kbps
N/A
Channel
Configuration
Default
Dolby Digital
(AC-3) Bit Rate
Default
Dolby Digital Plus
(E-AC-3) Bit Rate
Stereo (2/0)
192 kbps
128 kbps
5.1
384 kbps
192 kbps
