In general usage, one meaning of the wordscript is the written text of a film, television programme, play etc. A script can be either a record of the completed production, also known as a transcript, or as a plan for a yet to be created production. In this document, we use domain-specific terms, and define more specifically that:

• atranscript is the text representation ofpre-existing media in another form, for example the dialogue in a video;
• ascript is a text representation of theintended content of media prior to its creation, for example to guide an actor in recording an audio track.

The termDAPT script is used generically to refer to bothtranscripts andscripts, and is a point of conformance to the formal requirements of this specification.DAPT Scripts consist of timed text and associated metadata, such as the character speaking.

In dubbing workflows, atranscript is generated and translated to create ascript. In audio description workflows, atranscript describes the video image, and is then used directly as ascript for recording an audio equivalent.

DAPT is aTTML-based format for the exchange oftranscripts andscripts (i.e.DAPT Scripts) among authoring, prompting and playback tools in the localization andaudio description pipelines. ADAPT document is a serializable form of aDAPT Script designed to carry pertinent information for dubbing oraudio description such as type ofDAPT script, dialogue, descriptions, timing, metadata, original language transcribed text, translated text, language information, and audio mixing instructions, and to be extensible to allow user-defined annotations or additional future features.

This specification defines the data model forDAPT scripts and its representation as a [TTML2] document (see4.DAPT Data Model and correspondingTTML syntax) with some constraints and restrictions (see5.Constraints).

ADAPT script is expected to be used to make audio visual media accessible or localized for users who cannot understand it in its original form, and to be used as part of the solution for meeting user needs involvingtranscripts, including accessibility needs described in [media-accessibility-reqs], as well as supporting users who need dialogue translated into a different language via dubbing.

Every part of theDAPT script content is required to be marked up with some indication of what it represents in the related media, via theRepresents property; likewise theDAPT Script as a whole is required to list all the types of content that it represents, for example if it represents audio content or visual content, and if visual, then if it represents text or non-text etc. A registry of hierarchical content descriptors is provided.

The authoring workflow for both dubbing and audio description involves similar stages, that share common requirements as described in [DAPT-REQS]. In both cases, the author reviews the content and writes down what is happening, either in the dialogue or in the video image, alongside the time when it happens. Further transformation processes can change the text to a different language and adjust the wording to fit precise timing constraints. Then there is a stage in which an audio rendering of thescript is generated, for eventual mixing into the programme audio. That mixing can occur prior to distribution, or in the player directly.

ADAPT Script is atranscript orscript that corresponds to a document processed within an authoring workflow or processed by a client, and conforms to the constraints of this specification. It has properties and objects defined in the following sections:Script Represents,Script Type,Default Language,Text Language Source,Script Events and, forDubbing Scripts,Characters.

ADAPT Document is a [TTML2]timed text content document instance representing aDAPT Script. ADAPT Document has the structure and constraints defined in this and the following sections.

daptm:scriptRepresents: <content-descriptor> ( <lwsp>+ <content-descriptor>)*<lwsp>                # as TTML2

daptm:scriptType  : "originalTranscript"  | "translatedTranscript"  | "preRecording"  | "asRecorded"

<ttdaptm:scriptType="originalTranscript">...</tt>

<content-descriptor>  # see registry table below: <descriptor-token> ( <descriptor-delimiter> <descriptor-token> )*<descriptor-token>: (descriptorTokenChar)+descriptorTokenChar  #xsd:NMtoken without the ".":NameStartChar | "-" | [0-9] | #xB7 | [#x0300-#x036F] | [#x203F-#x2040]<descriptor-delimiter>: "."  # FULL STOP U+002E

This section is mainly relevant toDubbing workflows.

A character in the programme can be described using aCharacter object which has the following properties:

• a mandatoryCharacter Identifier which is aUnique Identifier used to reference the character from elsewhere in the document, for example to indicate when aCharacter participates in aScript Event.
• a mandatoryName which is the name of theCharacter in the programme
• an optionalTalent Name, which is the name of the actor speaking dialogue for thisCharacter.

ACharacter is represented in aDAPT Document by the following structure and constraints:

• TheCharacter is represented in aDAPT Document by a<ttm:agent> element present at the path/tt/head/metadata/ttm:agent, with the following constraints:
  • Thetype attributeMUST be set tocharacter.
  • Thexml:id attributeMUST be present on the<ttm:agent> element and set to theCharacter Identifier.
  • The<ttm:agent> elementMUST contain a<ttm:name> element with itstype attribute set toalias and its content set to theCharacter Name.

  • If the Character has aTalent Name, itMUST contain a<ttm:actor> child element. That child elementMUST have anagent attribute set to the value of thexml:id attribute of a separate<ttm:agent> element corresponding to theTalent Name, that is, whosetype attribute is set toperson.

    Note

    The requirement for an additional<ttm:agent> element corresponding to theTalent Name is defined in the following bullet list.

  Example 19

  ...<metadata><ttm:agenttype="character"xml:id="character_1"><ttm:nametype="alias">DESK CLERK</ttm:name></ttm:agent></metadata>...

  Example 20

  ...<metadata><ttm:agenttype="person"xml:id="actor_A"><ttm:nametype="full">Matthias Schoenaerts</ttm:name></ttm:agent><ttm:agenttype="character"xml:id="character_2"><ttm:nametype="alias">BOOKER</ttm:name><ttm:actoragent="actor_A"/></ttm:agent></metadata>...

• If theCharacter has aTalent Name property:
  • A<ttm:agent> element corresponding to theTalent NameMUST be present at the path/tt/head/metadata/ttm:agent, with the following constraints:
    • itstype attributeMUST be set toperson
    • itsxml:id attributeMUST be set.
    • itMUST have a<ttm:name> child element whosetypeMUST be set tofull and its content set to theTalent Name
  • If more than oneCharacter is associated with the sameTalent Name thereSHOULD be a single<ttm:agent> element corresponding to thatTalent Name, referenced separately by each of theCharacters.
  • Each<ttm:agent> element corresponding to aTalent NameSHOULD appear before any of theCharacter<ttm:agent> elements whose<ttm:actor> child element references it.
• All<ttm:agent> elementsSHOULD be contained in the first<metadata> element in the<head> element.
  Note

  There can be multiple<metadata> elements in the<head> element, for example to includeproprietary metadata but the above recommends that only one is used to define the characters.
  Editor's note

  The group is considering updating the rule(s) around which metadata element is used to carryDAPT information. The group would like to balance simplicity of implementation (e.g. locating theDAPT metadata in one place) vs. flexibility of authoring (e.g. having different metadata elements for series vs episodes). One approach is the current one: "only one metadata element, the first one". Another approach is "only one metadata element, identified by an attribute". Another approach is "any number of metadata elements". The group welcomes feedback from implementers and users.

AScript Event object represents dialogue, on screen text or audio descriptions to be spoken and has the following properties:

• A mandatoryScript Event Identifier which is aUnique Identifier.
• An optionalBegin property and an optionalEnd and an optionalDuration property that together define theScript Event's time interval in the programme timeline
  Note

  TypicallyScript Events do not overlap in time. However, there can be cases where they do, e.g. inDubbing Scripts when differentCharacters speak different text at the same time.

• ARepresents property used to identify what content the event represents (e.g. dialogue, on screen text, etc.). EveryScript EventMUST have a validRepresents property.
  Note

  TheRepresents property is inheritable, so the value does not need to be specified on everyScript Event explicitly.
• Zero or moreCharacter Identifiers indicating theCharacters involved in thisScript Event.
  Note

  While typically, aScript Event corresponds to one singleCharacter, there are cases where multiplecharacters can be associated with aScript Event. This is when allCharacters speak the same text at the same time.

  Note

  In atranscript, when the event corresponds to in-image content, for example anaudio description, noCharacter Identifier is needed. However it may be helpful in aPre-recording Script or anAs-recorded Script context to indicate aCharacter signifying who voices the recording.

• Zero or moreText objects, each being eitherOriginal orTranslation.

  Note

  AScript Event with noText objects can be created as part of an initial phase of authoring, in workflows where it is helpful to block out the time intervals during which some content could be present. For example, an emptyScript Event withtiming properties can be created to identify an opportunity for creating an audio description. See also[DAPT-REQS] Process Step 1.

  Note

  EmptyText objects, i.e. ones that have no text content, can be used to indicate explicitly that there is no text content. It is recommended that emptyText objects are not used as a workflow placeholder to indicate incomplete work.

• zero or moreScript Event Description objects, each being a human-readable description of theScript Event.
• an optionalOn Screen property, which is an annotation indicating the position of the subject (e.g. acharacter) of theScript Event
• Zero or moreMixing Instruction objects used to adjust playback of the programme audio during theScript Event.

AScript Event is represented in aDAPT Document at the path/tt/head/body//div, with the following structure and constraints:

• ThereMAY be any number of nested<div> element ancestors in the path between the<body> element and the<div> element corresponding to theScript Event. No further semantic is defined for such elements.
• ThereMUST be one<div> element corresponding to theScript Event, with the following constraints:
  • Thexml:id attributeMUST be present containing theScript Event Identifier.

  • Thebegin,end anddur attributes represent respectively theBegin,End andDuration of theScript Event.

    Thebegin andend attributesSHOULD be present. Thedur attributeMAY be present.

    Note

    See4.1.6.1Timing Properties for additional notes on timing properties.

  • Thettm:agent attributeMAY be present and if present,MUST contain a reference to eachttm:agent attribute that represents an associatedCharacter.
    Note

    Multiple references are specified using a space-separated list.
    Example 21

    ...<divxml:id="event_1"begin="9663f"end="9682f"ttm:agent="character_4">...</div>...

  • Thedaptm:represents attributeMAY be present representing theRepresents property.
    Example 22

    ...<divxml:id="event_1"begin="9663f"end="9682f"daptm:represents="audio.dialogue">...</div>...

  • The computed value of the thedaptm:represents attributeMUST be a valid non-empty value.
    Note

    It is possible to specify thedaptm:represents attribute on an ancestor element, sinceRepresents is an inheritable property.
  • ItMAY contain zero or more<p> elements representing eachText object.
  • ItMAY contain a<metadata> element representing theOn Screen property.
  • ItMUST NOT contain any<div> element children.

TheText object contains text content typically in a single language. This language may be theOriginal language or aTranslation language.

Text is defined asOriginal if it is any of:

1. the same language as the dialogue that it represents in the original programme audio;
2. a transcription of text visible in the programme video, in the same language as that text;
3. an untranslated representative of non-dialogue sound;
4. an untranslated description of the scene in the programme video.

Text is defined asTranslation if it is a representation of anOriginalText object in a different language.

Text can beidentified as beingOriginal orTranslation by inspecting its language and itsText Language Source together, according to the semantics defined inText Language Source.

The source language ofTranslationText objects and, where applicable,OriginalText objects is indicated using theText Language Source property.

AText object may be styled.

Zero or moreMixing Instruction objects used to modify the programme audio during theTextMAY be present.

AText object is represented in aDAPT Document by a<p> element at the path/tt/head/body//div/p, with the following constraints:

• TheText of theScript Event is represented by the character content of the<p> element and of all of its<span> descendant elements, after<metadata> elements and foreign elements have been pruned, after replacing<br> elements by line breaks, and after applying White Space Handling as defined in [XML].
  Note

  The text content of the paragraph can be structured usingTTML elements such as<br> or<span> which can include or referenceTTML style attributes such astts:ruby used to alter the layout or styling of sections of text within each paragraph. Mixed direction text, for example interleaved left to right (ltr) and right to left (rtl) text, can be specified by using thetts:direction attribute on<span> elements. Similarly metadata can be added using attributes or<metadata> elements.
• The<p> elementSHOULD have adaptm:langSrc attribute representing theText object'sText Language Source, that is, indicating whether theText isOriginal or aTranslation and if its source had an inherent language.
  Note

  If a<p> element omits thedaptm:langSrc attribute then its computed value is derived by inheritance from its parent element, and so forth up to the root<tt> element.

  Inscripts that have very little variation in source language, thedaptm:langSrc attribute can be set on the root element and omitted from<p> elements except where its value differs.

  Care should be taken if using this approach, especially when moving betweenScript Types, because changing it at the root element could affect the interpretation of descendant elements unexpectedly. In tools that allow fine-grained control, authors can mitigate this risk by explicitly setting thedaptm:langSrc attribute on all<p> elements.

  Implementers should take care to ensure that, when changing thedaptm:langSrc attribute on an element, they check down the tree and if appropriate specify the attribute on descendant elements so that their meaning does not change unintentionally.

  Note

  A document that does not specify thedaptm:langSrc attribute at all implies that all of the text is a transcript of content with no inherent language, for example audio description where no in-image text is transcribed, and which has not been translated.

• The<p> elementSHOULD have anxml:lang attribute corresponding to the language of theText object.
  Note

  If a<p> element omits thexml:lang attribute then its computed language is derived by inheritance from its parent element, and so forth up to the root<tt> element, which is required to set theDefault Language via itsxml:lang attribute.

  Care should be taken if changing theDefault Language of aDAPT Script in case doing so affects descendant elements unexpectedly. In tools that allow fine-grained control, authors can mitigate this risk by explicitly setting thexml:lang attribute on all<p> elements.

  Implementers should take care to ensure that, when changing thexml:lang attribute on an element, they check down the tree and if appropriate specify the attribute on descendant elements so that their meaning does not change unintentionally.

  Example 23

  <divxml:id="event_3"begin="9663f"end="9682f"ttm:agent="character_3"><pxml:lang="pt-BR">Você vai ter.</p><pxml:lang="fr"daptm:langSrc="pt-BR">Bah, il arrive.</p></div>

  Note

  In some cases, a single section of untranslated dialogue can contain text in more than one language. Rather than splitting aScript Event into multipleScript Events to deal with this,Text objects in one language can also contain some words in a different language. This is represented in aDAPT Document by setting thexml:lang anddaptm:langSrc attributes on inner<span> elements.

  Note

  <span> elements can be used to add specific timing as illustrated inExample 10 to indicate the timing of the audio rendering of the relevant section of text. Per [TTML2], timing of the<span> element is relative to the parent element's computed begin time.

• ItMAY contain zero or more<audio> elements representing eachAudio Recording object.
• ItMAY contain zero or more<animate> elements representing eachMixing Instruction object.

TheText Language Source property is an annotation indicating the source language of aText object, if applicable, or that the source content had no inherent language:

• If it is empty, theText represents content without an inherent language, such as untranslated descriptions of a visual scene or captions representing non-dialogue sounds.
• Otherwise (if it is not empty):
  • if its value is the same as the language of theText object, theText isOriginal.
  • otherwise (if the value is different to the language of theText), theText is aTranslation, whose source language is theText Language Source property's value.

Text Language Source is an inheritable property.

TheText Language Source property is represented in aDAPT Document by adaptm:langSrc attribute with the following syntax, constraints and semantics:

daptm:langSrc: <empty-string> | <language-identifier><empty-string>: ""                    # default<language-identifier>   # valid BCP-47 language tag

• The valueMUST be an empty string or a language identifier as defined by [BCP47].
• The default value is the empty string.
• Itapplies to<p> and<span> elements.
• ItMAY be specified on the following elements:<tt>,<p> and<span>.
• The inheritance model of thedaptm:langSrc attribute is as follows:
  • If it is present on an element, the computed value is the specified value.
  • Otherwise (if it is not present on an element), the computed value of the attribute on that element is the computed value of the same attribute on the element's parent, or if the element has no parent it is the default value.
  Note

  The inheritance model of thedaptm:langSrc attribute is intended to match the inheritance model of thexml:lang attribute [XML].

• The semantics of the computed value are as follows:
  • If the computed value is the empty string then it indicates that theText isOriginal and sourced from content without an inherent language.
  • Otherwise (the computed value is not empty)
    • if the computed value is the same as the computed value of thexml:lang attribute, then it indicates that theText isOriginal and sourced from content with an inherent language.
    • Otherwise (the computed value differs from the computed value of thexml:lang attribute), it indicates that theText is atranslation, and the computed value is the language from which theText was translated.

If any of these transcripts were translated, the resultingText would have itsdaptm:langSrc attribute set to the computed value of thexml:lang attribute of the source.

For example, if the Arabic dialogue were translated into Japanese, it would result inxml:lang="ja" anddaptm:langSrc="ar".

TheOn Screen property is an annotation indicating the position in the scene relating to the subject of aScript Event, for example of thecharacter speaking:

• ON - theScript Event's subject is on screen for the entire duration
• OFF - theScript Event's subject is off screen for the entire duration
• ON_OFF - theScript Event's subject starts on screen, but goes off screen at some point
• OFF_ON - theScript Event's subject starts off screen, but goes on screen at some point

If omitted, the default value is "ON".

TheOn Screen property is represented in aDAPT Document by adaptm:onScreen attribute on the<div> element, with the following constraints:

• The following attribute corresponding to theOn ScreenScript Event property may be present:

  daptm:onScreen  : "ON"     # default  | "OFF"  | "ON_OFF"  | "OFF_ON"

TheRepresents property indicates which component of therelated media object theScript Event represents.

TheRepresents property is represented in aDAPT Document by adaptm:represents attribute, whose valueMUST be a single<content-descriptor>.

TheRepresents property is inheritable. If it is absent from an element then its computed value is the computed value of theRepresents property on its parent element, or, if it has no parent element, it is the empty string. If it is present on an element then its computed value is the value specified.

It is an error for aRepresents property value not to be acontent descriptorsub-type of at least one of the values in theScript Represents property.

TheScript Event Description object is an annotation providing a human-readable description of some aspect of the content of aScript Event. Script Event Descriptions can themselves be classified with aDescription Type.

AScript Event Description object is represented in aDAPT Document by a<ttm:desc> element at the<div> element level.

Zero or more<ttm:desc> elementsMAY be present.

Script Event DescriptionsSHOULD NOT be empty.

The<ttm:desc> elementMAY specify its language using thexml:lang attribute.

...<body><divbegin="10s"end="13s"xml:id="a1"><ttm:desc>Scene 1</ttm:desc><pxml:lang="en"><span>A woman climbs into a small sailing boat.</span></p><pxml:lang="fr"daptm:langSrc="en"><span>Une femme monte à bord d'un petit bateau à voile.</span></p></div><divbegin="18s"end="20s"xml:id="a2"><ttm:desc>Scene 1</ttm:desc><pxml:lang="en"><span>The woman pulls the tiller and the boat turns.</span></p><pxml:lang="fr"daptm:langSrc="en"><span>La femme tire sur la barre et le bateau tourne.</span></p></div></body>...

EachScript Event Description can be annotated with one or moreDescription Types to categorise further the purpose of the Script Event Description.

EachDescription Type is represented in aDAPT Document by adaptm:descType attribute on the<ttm:desc> element.

The<ttm:desc> elementMAY have zero or onedaptm:descType attributes. Thedaptm:descType attribute is defined below.

daptm:descType : string

The permitted values fordaptm:descType are either those listed in the followingregistry table, or can be user-defined:

Valid user-defined valuesMUST begin withx-.

...<body><divbegin="10s"end="13s"xml:id="a123"><ttm:descdaptm:descType="pronunciationNote">[oːnʲ]</ttm:desc><p>Eóin looks around at the other assembly members.</p></div></body>...

Amongst a sibling group of<ttm:desc> elements there are no constraints on the uniqueness of thedaptm:descType attribute, however it may be useful as a distinguisher as shown in the following example.

...<body><divbegin="10s"end="13s"xml:id="a1"><ttm:descdaptm:descType="scene">Scene 1</ttm:desc><ttm:descdaptm:descType="plotSignificance">High</ttm:desc><pxml:lang="en"><span>A woman climbs into a small sailing boat.</span></p><pxml:lang="fr"daptm:langSrc="en"><span>Une femme monte à bord d'un petit bateau à voile.</span></p></div><divbegin="18s"end="20s"xml:id="a2"><ttm:descdaptm:descType="scene">Scene 1</ttm:desc><ttm:descdaptm:descType="plotSignificance">Low</ttm:desc><pxml:lang="en"><span>The woman pulls the tiller and the boat turns.</span></p><pxml:lang="fr"daptm:langSrc="en"><span>La femme tire sur la barre et le bateau tourne.</span></p></div></body>...

AnAudio object is used to specify an audio rendering of aText. The audio rendering can either be a recorded audio resource, as anAudio Recording object, or a directive to synthesize a rendering of the text via a text to speech engine, which is aSynthesized Audio object. Both are types ofAudio object.

It is an error for anAudio not to be in the same language as itsText.

Apresentation processor that supports audio plays or inserts theAudio at the specified time on therelated media object's timeline.

AMixing Instruction object is a static or animated adjustment of the audio relating to the containing object. It has the following properties:

• Zero or moreGain properties. The gain acts as a multiplier to be applied to the relatedAudio;
• Zero or morePan properties. The pan adjusts the stereoscopic (left/right) position;
• An optionalBegin and an optionalEnd and an optionalDuration property that together define the time interval during which theMixing Instruction applies;
• An optionalFill property that specifies whether, at the end time of an animatedMixing Instruction, the specifiedGain andPan properties should be retained (freeze) or reverted (remove).

AMixing Instruction is represented by applying audio style attributes to the element that corresponds to the relevant object, either inline, by reference to a<style> element, or in a child (inline)<animate> element:

• Thetta:gain attribute represents theGain property;
• Thetta:pan attribute represents thePan property.

If theMixing Instruction is animated, that is, if the adjustment properties change during the containing object's active time interval, then it is represented by one or more child<animate> elements. This representation is required if more than oneGain orPan property is needed, or if anytiming properties are needed.

The<animate> element(s)MUST be children of the element corresponding to the containing object, and have the following constraints:

• Thebegin,end anddur attributes represent respectively theBegin,End andDuration properties;
• Thefill attribute represents theFill property;
• Thetta:gain attribute represents theGain property, and uses theanimation-value-list syntax to express the list of values to be applied during the animation period;
• Thetta:pan attribute represents thePan property, and uses theanimation-value-list syntax to express the list of values to be applied during the animation period.

TheTTML representation of animatedMixing Instructions is illustrated byExample 4.

See alsoE.Audio Mixing.

ADAPT DocumentMUST be serialised as a well-formed XML 1.0 [xml] document encoded using the UTF-8 character encoding as specified in [UNICODE].

The resulting [xml] documentMUST NOT contain any of the following physical structures:

• entity declarations; and
• entity references other than topredefined entities.

The requirements in this section are intended to facilitate forwards and backwards compatibility, specifically to permit:

• DAPT processors targeted at one version of the specification to processDAPT documents that include vocabulary or semantics defined in future versions, albeit without supporting the later features;
• DAPT processors targeted at one version of the specification to processDAPT documents authored for an earlier version with similar or identical behaviour to aDAPT processor targeted at that earlier version.

ADAPT document that conforms to more than one version of the specification could specify conformance to multipleDAPTcontent profiles.

...<head><metadata><ttm:title>A example document title</ttm:title></metadata></head>...

...<metadataxmlns:vendorm="http://www.example-vendor.com/ns/ttml#metadata"><vendorm:programType>Episode</vendorm:programType><vendorm:episodeSeason>5</vendorm:episodeSeason><vendorm:episodeNumber>8</vendorm:episodeNumber><vendorm:internalId>15734</vendorm:internalId><vendorm:information>Some proprietary information</vendorm:information></metadata>...

The following namespaces (see [xml-names]) are used in this specification:

The namespace prefix values defined above are for convenience andDAPT DocumentsMAY use any prefix value that conforms to [xml-names].

The namespaces defined by this specification are mutable as described in [namespaceState]; all undefined names in these namespaces are reserved for future standardization by theW3C.

WithinDAPT, the common language terms audio and video are used in the context of a programme. The audio and video are each a part of what is defined in [TTML2] as theRelated Media Object that provides the media timeline and is the source of themain programme audio, and any visual timing references needed when adjusting timings relevant to the video image, such as for lip synchronization.

<ttxmlns="http://www.w3.org/ns/ttml"xmlns:ttp="http://www.w3.org/ns/ttml#parameter"xmlns:daptm="http://www.w3.org/ns/ttml/profile/dapt#metadata"xmlns:ebuttm="urn:ebu:tt:metadata"xml:lang="en"ttp:contentProfiles="http://www.w3.org/ns/ttml/profile/dapt1.0/content"daptm:scriptRepresents="audio.dialogue"daptm:scriptType="originalTranscript"><head><metadata><ebuttm:sourceMediaIdentifier>https://example.org/programme.mov</ebuttm:sourceMediaIdentifier></metadata></head><body>    ...</body></tt>

If theDAPT Document is intended to be used as the basis for producing an [TTML-IMSC1.2] document, the synchronization provisions of [TTML-IMSC1.2] apply in relation to the video.

Timed content within theDAPT Document is intended to be rendered starting and ending on specific audio samples.

In constrained applications, such as real-time audio mixing and playback, if accurate synchronization to the audio sample cannot be achieved in the rendered output, the combined effects of authoring and playback inaccuracies in timed changes in presentationSHOULD meet the synchronization requirements of [EBU-R37], i.e. audio changes are not to precede image changes by more than 40ms, and are not to follow them by more than 60ms.

Likewise, authoring applicationsSHOULD allow authors to meet the requirements of [EBU-R37] by defining times with an accuracy such that changes to audio are less than 15ms after any associated change in the video image, and less than 5ms before any associated change in the video image.

Taken together, the above two constraints on overall presentation and onDAPT documents intended for real-time playback mean thatcontent processorsSHOULD complete audio presentation changes no more than 35ms before the time specified in theDAPT document and no more than 45ms after the time specified.

Within aDAPT Script, the following constraints apply in relation to time attributes and time expressions:

This specification does not put additional constraints on the layout and renderingfeatures defined in [TTML-IMSC1.2].

Style references or inline stylesMAY be used, using any combination ofstyle attributes,<style> elements and inline style attributes as defined in [TTML2] or [TTML-IMSC1.2].

The following metadata elements are permitted inDAPT and specified in [TTML2] as containing#PCDATA, i.e. text data only with no element content. Where bidirectional text is required within the character content within such an element, Unicode control characters can be used to define the base direction within arbitrary ranges of text.

• <ttm:copyright>
• <ttm:desc>
• <ttm:item>
• <ttm:name>
• <ttm:title>

The<p> and<span> content elements permit the direction of text to be specified using thetts:direction andtts:unicodeBidi attributes. Document authors should use this more robust mechanism rather than using Unicode control characters.

Note

The following example taken from [TTML2] demonstrates the syntax for bidirectional text markup within the<p> and<span> elements.

Example 36

<p>The title of the book is"<spantts:unicodeBidi="embed"tts:direction="rtl">نشاط التدويل، W3C</span>"</p>

An example rendering of the above fragment is shown below.

This section is non-normative.

Normative provisions relating to this section are defined in [TTML2].

Since it is a requirement ofDAPT thatDAPT Documents include attp:contentProfiles attribute on the root element, and that the attribute includes aDAPTcontent profile designator, as specified at5.6.2ttp:contentProfiles, it follows that anyTTML document that does not include such an attribute, or does not include such a profile designator, can be considered not to be aDAPT Document; therefore a processor requiring strict adherence toDAPT could stop processing such a document.

A processor that takes as its input aDAPT document that contains vocabulary relating to features that it does support, but where support for those features is excluded from the content profiles to which the document claims conformance,SHOULD NOT implement those features in the context of that document.

[TTML2] allows<div> elements to contain any combination of<div> elements and<p> elements. TheDAPT data model describes how eachScript Event is represented by a<div> element that contains zero or more<p> elements. It also permits other intermediate<div> elements in the path between the<body> element and thoseScript Event<div> elements. In addition, attributes not corresponding to properties in theDAPT data model are permitted.

This gives rise to possibilities such as:

• <div> elements that contain children that are notText objects, for example other<div> elements or<metadata> elements;
• <div> elements that do not have attributes required forScript Events, likexml:id;
• <div> elements that have attributes absent from theDAPT data model forScript Events, likexml:space;
• <p> elements whose parent<div> element does not appear to be aScript Event.

The following processing rules resolve these cases.

Rules for identifyingScript Events:

1. A<div> element that has no<div> element children and includes theTTML representations of all the mandatory properties of aScript EventMUST be mapped to aScript Event, such as having a validxml:id representing theScript Event Identifier, even if it also contains additionalunrecognised vocabulary;
2. A<div> element that contains any<div> element childrenMUST NOT be mapped to aScript Event; the processor insteadMUST iterate through those<div> element children (recursively, in a depth-first traversal) and consider if each one meets the requirements of aScript Event;
3. Any remaining unmapped<div> elementsMUST NOT be mapped to aScript Event.

Rules for identifyingText objects:

1. A<p> element that is a child of a<div> element that maps to aScript EventMUST be mapped to aText object.
2. A<p> element that is not a child of a<div> element that maps to aScript EventMUST NOT be mapped to aText object.

<body><divxml:id="d1"><!-- This is a Script Event --><p><!-- This is a Text --></p></div><div><!-- This cannot be a Script Event because it has no xml:id --><p><!-- Would be a Text if its parent were a Script Event --></p></div><divxml:id="d2_1"><!-- div parent of another div --><divxml:id="d2"><!-- Possibly a Script Event --></div></div><divxml:id="d3_1"><!-- double layer of nesting --><divxml:id="d3_1_1"><divxml:id="d3"begin="..."end="..."xml:lang="ja"foo:bar="baz"><!-- A Script Event with possibly unexpected attributes --></div></div></div><divxml:id="d4_1"><!-- mixed div and p children --><divxml:id="d4_2"><!-- This possible Script Event has a sibling <p> --></div><p><!-- Possible Text, but not if its parent is not a Script Event --></p></div></body>

The result of applying the above rules to this example is:

• d1 is aScript Event and its child<p> element is aText object;
• The<div> element betweend1 andd2's parent<div> element is not aScript Event and its child<p> element is not aText object;
• d2 is aScript Event and its parent<div> element is not;
• d3 is aScript Event and its parent and its grandparent<div> elements are not;
• d4_1 is not aScript Event and its child<p> element is not aText object;
• d4_2 is aScript Event.

Some attributes have semantics for computing their value that depend on the computed value of the attribute on some other element. For example if thexml:lang attribute is not specified on an element then its computed value is the computed value of the same attribute on the element's parent.

Or, for another example, the computed times of an element in aDAPT document are relative to the begin time of the element's parent. If a<div> element specifies abegin attribute, then the computed times of its child<div> elements are relative to that parent<div> element's begin time, and so on down the hierarchy. It is important to include those "intermediate"<div> elements' times in the computation even if the processing target is an instance of theDAPT data model in which they have no direct equivalent; otherwise theScript EventBegin andEnd times would be wrong.

Considering this situation more generally, it is possible that, within aDAPT document, there can beTTML elements that do not directly correspond to objects in theDAPT data model, and those elements can specify attributes that affect the computation of attribute values that apply to elements that do correspond to objects in theDAPT data model.

The semantics defined by [TTML2] or, for vocabulary defined herein, this specification, take precedence in this scenario. ImplementationsMUST compute attribute values based on the contents of thedocument instancebefore applying those computed values toDAPT data model objects. For example a processor that supports TTML2 styling features would need to implement the TTML2 semantics for inheritance and computing relative values of attributes liketts:fontSize.

This example demonstrates these possibilities, which a processor might need to handle:

<tt...xml:lang="en"><body><divxml:id="d1"begin="00:01:00"end="00:01:10"><!-- Script Event beginning at 1 minute, duration 10 seconds --><p><!-- This is a Text, language "en" --></p></div><divbegin="00:10:00"xml:lang="fr"><!-- div that is not a Script Event --><divxml:id="d2"begin="00:01:00"end="00:01:10"><!-- Script Event beginning at 11 minutes, duration 10 seconds --><p><!-- This is a Text, language "fr" --></p></div></div></body></tt>

Here, the<div> elements that correspond toScript Events, which have idsd1 andd2, are otherwise (apart from their identifiers) identical, yet their language andBegin properties have different computed values, due to the attributes specified on the<div> element parent ofd2.

If an implementation internally createsDAPT data model objects, such as theScript Events shown in the example above, it is important for interoperability that they use the computed values.

This section is non-normative.

ConformantDAPT Documents are [TTML2]timed text content document instances that conform to the normative provisions of this specification. Those provisions are expressed using theprofile vocabulary of [TTML2] in thecontent profile defined inF.Profiles.

ConformantDAPT Processors are [TTML2]content processors that conform to the normative provisions of this specification. Those provisions are expressed using theprofile vocabulary of [TTML2] in theprocessor profile defined inF.Profiles.

• As-recorded Script§4.1.3
• Audio§4.9
• audio description§2.1.2
• Audio Recording§4.9.1
• Begin§4.1.6.1
• change request§H.1.2.1
• Character§4.2
• Character Identifier§4.2
• content descriptor sub-type§4.1.6.2
• content-descriptor§4.1.6.2
• DAPT 1.0 Content Profile§5.6.1
• DAPT 1.0 Processor Profile§5.6.1
• DAPT Document§4.1
• DAPT Extension§5.3
• DAPT Origin Timecode§D.1
• DAPT Script§4.1
• daptm:daptOriginTimecode§D.1
• daptm:descType§4.8
• daptm:langSrc§4.5
• daptm:onScreen§4.6
• daptm:represents§4.7
• daptm:scriptRepresents§4.1.1
• daptm:scriptType§4.1.3
• Default Language§4.1.2
• Deprecated§H.1.3.1.1.3
• description§2.1.2
• Description Types§4.8
• descriptor-token§4.1.6.2
• dubbing§2.1.1
• dubbing script§2.1.1
• Duration§4.1.6.1
• End§4.1.6.1
• fields§H.1.3.1
• Fill§4.10
• Final§H.1.3.1.1.2
• Foreign vocabulary§5.2.2
• Gain§4.10
• In Time§4.9.1
• key§H.1.3.1
• Main programme audio§2.1.2
• Mixing Instruction§4.10
• Name§4.2
• non-matching§G.17
• On Screen§4.6
• Original§4.4
• Original Language Transcript§4.1.3
• Out Time§4.9.1
• Pan§4.10
• Pitch§4.9.2
• Pre-recording Script§4.1.3
• Provisional§H.1.3.1.1.1
• Rate§4.9.2
• registry table definition§H.1.3
• Represents§4.7
• script§2.1
• Script Event§4.3
• Script Event Description§4.8
• Script Event Identifier§4.3
• Script Represents§4.1.1
• Script Type§4.1.3
• shared properties§4.1.6
• Sources§4.9.1
• Start of Programme Timecode§D.2
• status§H.1.3.1.1
• Synthesized Audio§4.9.2
• Talent Name§4.2
• Target Recording Language§4.1.2
• Text§4.4
• Text Language Source§4.5
• Timed Text Working Group§H.1.1
• timing properties§4.1.6.1
• transcript§2.1
• Translated Transcript§4.1.3
• Translation§4.4
• Type§4.9.1
• Unique Identifier§4.1.6.3
• Unrecognised vocabulary§5.2.1

• [DAPT-REQS] defines the following:
  • [DAPT-REQS] Process Step 1
• [I18N-INLINE-BIDI] defines the following:
  • Inline markup and bidirectional text in HTML
• [MEDIA-ACCESSIBILITY-REQS] defines the following:
  • described video
• [TTML2] defines the following:
  • #animate-fill
  • #animate-minimal
  • #animation-out-of-line
  • #audio
  • #audio-description
  • #audio-speech
  • #bidi
  • #bidi-version-2
  • #chunk
  • #clockMode
  • #clockMode-gps
  • #clockMode-local
  • #clockMode-utc
  • #content
  • #contentProfiles
  • #contentProfiles-combined
  • #core
  • #data
  • #direction
  • #dropMode
  • #dropMode-dropNTSC
  • #dropMode-dropPAL
  • #dropMode-nonDrop
  • #embedded-audio
  • #embedded-data
  • #frameRate
  • #frameRateMultiplier
  • #gain
  • #markerMode
  • #markerMode-continuous
  • #markerMode-discontinuous
  • #metadata
  • #metadata-item
  • #metadata-version-2
  • #pan
  • #permitFeatureNarrowing
  • #permitFeatureWidening
  • #pitch
  • #presentation-audio
  • #processorProfiles
  • #processorProfiles-combined
  • #profile
  • #profile-full-version-2
  • #profile-version-2
  • #resources
  • #set
  • #set-fill
  • #set-multiple-styles
  • #source
  • #speak
  • #speech
  • #structure
  • #styling
  • #styling-chained
  • #styling-inheritance-content
  • #styling-inline
  • #styling-referential
  • #subFrameRate
  • #tickRate
  • #time-clock
  • #time-clock-with-frames
  • #time-offset
  • #time-offset-with-frames
  • #time-offset-with-ticks
  • #time-wall-clock
  • #timeBase-clock
  • #timeBase-media
  • #timeBase-smpte
  • #timeContainer
  • #timing
  • #transformation
  • #unicodeBidi
  • #unicodeBidi-isolate
  • #unicodeBidi-version-2
  • #xlink
  • <lwsp>
  • <time-expression>
  • <ttp:extension>
  • <ttp:feature>
  • <ttp:profile>
  • content processors
  • content profile
  • Document Instance
  • extensions
  • features
  • presentation processor
  • processor profile
  • Profile
  • related media object
  • timed text content document instance
  • transformation processor
  • ttp:contentProfiles
  • ttp:processorProfiles
  • ttp:profile
  • validation processor
• [W3C-PROCESS] defines the following:
  • at risk
  • closed
  • custodian
  • registry definition
  • registry entries
  • registry section
  • registry tables
  • Updating Registry Tables
  • w3c registries
  • W3C Team
  • wide review
• [XML] defines the following:
  • character references
  • entity references
  • Name
  • NameStartChar
  • predefined entities
• [XMLSCHEMA-2] defines the following:
  • xsd:NMtoken
• [XPATH] defines the following:
  • LocationPath

DAPT documents typically contain the names of characters or people who feature within the associated media, either fictional or real. In general this information would be present within the media itself or be public via other routes. If there is sensitivity associated with their being known to people with access to theDAPT documents in which their identity is contained, then such access should be managed with appropriate confidentiality. For example those documents could be available within a closed authoring environment and edited to remove the sensitive information prior to distribution to a wider audience. If this scenario arises, information security good practices within the closed environment should be applied, such as encryption of the document "at rest" and when being moved, access controlled by authentication platforms, etc.

DAPT documents can reference a set of alternate external audio resources for the same fragment of audio, where the processor is expected to select one of the alternatives based on features such as format support. If this pattern is used, it is possible that the processor's choice of audio resource, being exposed to the origin, reveals information about that processor, such as its preferred audio format.

The optionalDAPT Origin Timecode allows a timecode value to be declared that corresponds to the zero point of theDAPT document timeline, that is, the time of a hypotheticalScript Event whoseBegin is zero seconds.

The properties can be used to provide timecodes from the related media object or from any other script format used to produce theDAPT document, and are informational. However, when they are both provided and differ, it is an indication that theDAPT document is not synchronized with the related media object and that processing of the script event begins is needed. To achieve synchronization, the following needs to be done:

1. The differenceDAPT Origin Timecode minus theStart of Programme Timecode is computed as "delta". It may be positive or negative.
2. EachScript Event'sBegin andEnd value X shall be changed to X + delta.
3. TheDAPT Origin Timecode shall be removed or changed to theStart of Programme timecode value so that if this algorithm is run again it will result in delta that is zero.

ADAPT ScriptMAY contain zero or oneDAPT Origin Timecode objects.

TheDAPT Origin Timecode object is represented in aDAPT Document by a<daptm:daptOriginTimecode> element present at the path/tt/head/metadata/daptm:daptOriginTimecode, with the following constraints:

• The character content of the element must conform to aclock-time with aframes component, as defined in<time-expression>.

If the related media object contains a timecode for the video frame synchronized to the origin of theDAPT document timeline, theDAPT origin timecode is equal that timecode.

...<head><metadata><daptm:daptOriginTimecode>10:01:20:12</daptm:daptOriginTimecode></metadata></head><body><divxml:id="se1"begin="0s"end="1.8s"><!-- This script event was generated from a source whose begin timecode was 10:01:20:12 --></div></body>...

When converting legacy formats that store the equivalent of each Script Event, for example adescription, with SMPTE timecodes, but where theStart of Programme Timecode is not stored, and is not immediately available, theDAPT Origin Timecode can be used to defer synchronisation with the media until that media begin timecode is known.

One approach to deferring synchronisation usingDAPT Origin Timecode is as follows, assuming that a legacy source file is being converted to aDAPT Document and relates to media with continuous timecode:

• Find the earliest begin timecode for an item in the source file that will be mapped to aScript Event;
• Create aDAPT Document with aDAPT Origin Timecode set to that earliest begin timecode;
• For each relevant item in the source file, create aScript Event with appropriateBegin andEnd times obtained by converting the source file's begin and end timecodes to media time, relative to thatDAPT Origin Timecode, basing elapsed time on the applicable frame rate;
• If using frame-based media time expressions, set thettp:frameRate attribute of the<tt> element to the applicable frame rate;

When the related media's timecode is later dicovered, the times of theScript Events can be adjusted so that media time zero is coincident with theStart of Programme Timecode, and theDAPT Origin Timecode can be removed or set to the now-known value.

The optionalStart of Programme Timecode allows a timecode value to be declared that corresponds to the beginning of the related media object's programme content.

In combination withDAPT Origin Timecode, the value ofStart of Programme Timecode can be used to infer whether or not the media times in theDAPT Script are likely to be correctly synchronised with theRelated Media Object.

If bothDAPT Origin Timecode andStart of Programme Timecode are present, but their values are different, it is likely that the media times are not synchronised with theRelated Media Object, since this implies that the equivalent time code to zero seconds in media time is not the start of the programme, which is the requirement for correctly synchronised media time.

ADAPT ScriptMAY contain zero or oneStart of Programme Timecode objects.

TheStart of Programme Timecode object is represented in aDAPT Document by a<ebuttm:documentStartOfProgramme> element present at the path/tt/head/metadata/ebuttm:documentStartOfProgramme, with the syntactic constraints as defined in [EBU-TT-3390].

...<head><metadata><daptm:daptOriginTimecode>10:01:20:12</daptm:daptOriginTimecode><ebuttm:documentStartOfProgramme>10:00:00:00<ebuttm:documentStartOfProgramme><!-- It is likely that this document is 1 minute, 20 seconds and 12 frames too early,  compared to the related media --></metadata></head><body><divxml:id="se1"begin="0s"end="1.8s"><!-- This script event was generated from a source whose begin timecode was 10:01:20:12 --></div></body>...

If the times of theScript Events are adjusted to bring the media time into synchronisation with theRelated Media Object, as noted inDAPT Origin Timecode, theStart of Programme TimecodeSHOULD NOT be changed, since it is an invariant feature of theRelated Media Object, and does not describe the times in theDAPT Document.

Thefeatures andextensions listed in this section express the minimal requirements forDAPT Documents,Presentation Processors, andTransformation Processors.DAPT DocumentsMAY additionally conform to other profiles, and include syntax not prohibited by theDAPTcontent profile.Presentation Processors andTransformation ProcessorsMAY support additional syntax and semantics relating to other profiles.

TheDAPT Content Profile expresses the conformance requirements ofDAPT Scripts using the profile mechanism of [TTML2]. It can be used by a validating processor that supports theDAPT Processor Profile to validate aDAPT Document.

There is no requirement to include theDAPT Content Profile within aDAPT Document.

<?xml version="1.0" encoding="utf-8"?><!-- this file defines the "dapt-content" profile of ttml --><profilexmlns="http://www.w3.org/ns/ttml#parameter"designator="http://www.w3.org/ns/ttml/profile/dapt1.0/content"combine="mostRestrictive"type="content"><featuresxml:base="http://www.w3.org/ns/ttml/feature/"><!-- required (mandatory) feature support --><featurevalue="required">#structure</feature><featurevalue="required">#timeBase-media</feature><!-- optional (voluntary) feature support --><featurevalue="optional">#animate-fill</feature><featurevalue="optional">#animate-minimal</feature><featurevalue="optional">#audio</feature><featurevalue="optional">#audio-description</feature><featurevalue="optional">#audio-speech</feature><featurevalue="optional">#bidi</feature><featurevalue="optional"extends="#bidi">#bidi-version-2</feature><featurevalue="optional">#chunk</feature><featurevalue="optional">#content</feature><featurevalue="optional">#contentProfiles</feature><featurevalue="optional">#contentProfiles-combined</feature><featurevalue="optional">#core</feature><featurevalue="optional">#data</feature><featurevalue="optional">#direction</feature><featurevalue="optional">#embedded-audio</feature><featurevalue="optional">#embedded-data</feature><featurevalue="optional">#frameRate</feature><featurevalue="optional">#frameRateMultiplier</feature><featurevalue="optional">#gain</feature><featurevalue="optional">#metadata</feature><featurevalue="optional">#metadata-item</feature><featurevalue="optional"extends="#metadata">#metadata-version-2</feature><featurevalue="optional">#pan</feature><featurevalue="optional">#permitFeatureNarrowing</feature><featurevalue="optional">#permitFeatureWidening</feature><featurevalue="optional">#pitch</feature><featurevalue="optional">#presentation-audio</feature><featurevalue="optional">#processorProfiles</feature><featurevalue="optional">#processorProfiles-combined</feature><featurevalue="optional">#resources</feature><featurevalue="optional"extends="#animation">#set</feature><featurevalue="optional">#set-fill</feature><featurevalue="optional">#set-multiple-styles</feature><featurevalue="optional">#source</feature><featurevalue="optional">#speak</feature><featurevalue="optional">#speech</feature><featurevalue="optional">#styling</feature><featurevalue="optional">#styling-chained</feature><featurevalue="optional">#styling-inheritance-content</feature><featurevalue="optional">#styling-inline</feature><featurevalue="optional">#styling-referential</feature><featurevalue="optional">#tickRate</feature><featurevalue="optional">#time-clock</feature><featurevalue="optional">#time-offset</feature><featurevalue="optional">#time-offset-with-frames</feature><featurevalue="optional">#time-offset-with-ticks</feature><featurevalue="optional">#timing</feature><featurevalue="optional">#transformation</feature><featurevalue="optional">#unicodeBidi</feature><featurevalue="optional">#unicodeBidi-isolate</feature><featurevalue="optional"extends="#unicodeBidi">#unicodeBidi-version-2</feature><featurevalue="optional">#xlink</feature><!-- prohibited feature support --><featurevalue="prohibited">#animation-out-of-line</feature><featurevalue="prohibited">#clockMode</feature><featurevalue="prohibited">#clockMode-gps</feature><featurevalue="prohibited">#clockMode-local</feature><featurevalue="prohibited">#clockMode-utc</feature><featurevalue="prohibited">#dropMode</feature><featurevalue="prohibited">#dropMode-dropNTSC</feature><featurevalue="prohibited">#dropMode-dropPAL</feature><featurevalue="prohibited">#dropMode-nonDrop</feature><featurevalue="prohibited">#markerMode</feature><featurevalue="prohibited">#markerMode-continuous</feature><featurevalue="prohibited">#markerMode-discontinuous</feature><featurevalue="prohibited">#subFrameRate</feature><featurevalue="prohibited">#time-clock-with-frames</feature><featurevalue="prohibited">#time-wall-clock</feature><featurevalue="prohibited">#timeBase-clock</feature><featurevalue="prohibited">#timeBase-smpte</feature><featurevalue="prohibited">#timeContainer</feature></features><extensionsxml:base="http://www.w3.org/ns/ttml/profile/dapt/extension/"><!-- required (mandatory) extension support --><extensionvalue="required">#contentProfiles-root</extension><extensionvalue="required">#represents</extension><extensionvalue="required">#scriptRepresents</extension><extensionvalue="required">#scriptType-root</extension><extensionvalue="required">#serialization</extension><extensionvalue="required">#xmlId-div</extension><extensionvalue="required">#xmlLang-root</extension><!-- optional (voluntary) extension support --><extensionvalue="optional">#agent</extension><extensionvalue="optional">#daptOriginTimecode</extension><extensionvalue="optional">#descType</extension><extensionvalue="optional">#onScreen</extension><extensionvalue="optional">#scriptEventGrouping</extension><extensionvalue="optional">#scriptEventMapping</extension><extensionvalue="optional">#textLanguageSource</extension><!-- prohibited extension support --><extensionvalue="prohibited">#profile-root</extension><extensionvalue="prohibited">#source-data</extension><extensionvalue="prohibited">#xmlLang-audio-nonMatching</extension></extensions></profile>