BREAK CAPTION GROUPS AT LOGICAL PLACES

Deciding when to end one caption group and create a new one requires paying attention to

three things:

1. The rhythm of speech. When a speaker pauses, that's a good spot to break.

2. The grammatical structure of the speech. Punctuation, conjunctions, and prepositions are

good spots to break.

3. The length (character count) of a caption group. You cannot exceed 60 characters in a

caption group.

Always create a new caption group whenever the speaker changes or a sentence ends with

punctuation . ? !

Caption groups must be under 60 characters.

A caption group can have a maximum of 60 characters. Dash has built-in color coding to help guide

you. Aim for the caption group box to be green or yellow. If the caption group box turns red, it is too

long.

If a sentence is longer than 60 characters, break mid-sentence. Breaking shouldn't interrupt

comprehension or readability, so you should try to break:

 After punctuation , : ;

 Or before conjunctions such as: and, nor, but, or, yet, so, by.

 Or before prepositional phrases such as: that, who, because, in order to, not only, as we, in

which, where, with, what, how, for, through, until, to, as, of.

 Or before complete proper nouns (e.g. do not break between "United States of America").
SPEAKER LABELING

Speaker ID Rules

1. Use a dash and a space EVERY time a NEW speaker starts speaking or when the speaker

CHANGES. I.e. "- "

o Atmospherics do not count as a change of speaker. If an atmospheric is used that

breaks up dialogue from the same speaker, do not include a dash after the atmospheric.

2. If the speaker CANNOT be visually identified, identify the speaker with a speaker ID. E.g., "-

[Mark]", "- [Narrator]"

o If a speaker is unknown, use an appropriate identifier. Some possible examples

could be "- [Interviewer]", "- [Guest]", "- [Ghostly Voice]"

o Exception: There is no need to identify a speaker if they are visible for any portion of

their dialogue AND they are not interrupted

 E.g. Mark starts speaking off-screen but then walks into the frame while

talking

ADDITIONAL NOTES

When you NEED to identify a speaker, here's how:

 You should use the speaker's first name if it is known: "- [Mark]"

 If a character name is not known, use a visible descriptive identifier: "- [Blonde Woman]"

o Never use race or other discriminatory identifiers. Instead, use a descriptor such as

occupation, clothing, height, etc.

ATMOSPHERICS

Include atmospherics for these main conditions:

1. A sound effect is heard which is integral to the story or message of a video. If a character

reacts to a sound, you should probably include it. E.g. "(gun fires)"

2. Background music is heard in a way that sets a specific mood as part of the story telling.

Only include a background music atmospheric if there's a significant gap in speech and the

music seems important. E.g. "(dramatic orchestral music)"

Sound effects often help tell the story

 Atmospherics are put in parentheses and are always in lowercase, E.g. "(loud snoring)".

o Atmospheric-only caption groups do not need a dash or speaker ID.

 Only include significant sound effects that help tell the story. Use your best judgment.

 If in doubt, include the atmospheric.

o E.g. If a character reacts to an off-screen gun shot, use "(gun fires)"

o E.g. If there is a group of children playing, you could use "(children laughing)"

o E.g. "(plane flying overhead)" or "(car honking)" should be included if characters

react to it.

 Include sounds made by the speaker, E.g. "(laughs loudly)".

 Atmospherics should always be present tense, E.g. "(laughs loudly)", never past tense

"(laughed loudly)"

 Always describe with an action verb, E.g. "(frogs croaking)", never with an onomatopoeia of

a sound "(ribbit, ribbit, ribbit)"

Mood Music

Music is often used in videos to help set a mood or underscore actions.

If there's at least a 2-second gap in speech AND it does not seem that the lyrics are intended to be

clearly heard AND the background music is setting a specific mood, then caption the atmospheric as

mood music.

 Most of the time you won't know the artist or title, so you should use a description:

o E.g. "(gentle music)", "(bright pop music)", "(heavy metal music)", "(electronic dance

music)".

o A list of music adjectives can be found here.

 If someone talks over the music, focus on the speaker and don't add the atmospheric.

Dialogue is always more important.

PRE-EXISTING ON-SCREEN TEXT

There are 2 main scenarios in which you will see pre-existing on-screen text:

1. Pre-existing text shown as slides, graphics, whiteboards, or a software interface. DO NOT

use the up-arrow caret ^.

2. Pre-existing text shown that helps tell the story AND is in the lower 1/3 of the screen. DO use

the up-arrow caret ^ to indicate the overlap.

How to indicate overlap with on-screen storytelling text in lower 1/3 of the screen:

1. Up-arrow carets ^ should be used whenever there is pre-existing burned-in text on the lower

1/3 of the screen.

2. When you flag this occurence, insert the up-arrow caret ^ in the caption group by typing

shift+6 on your keyboard and Dash will display a blue up-arrow caret at the beginning of the

caption group.

3. You must add an up-arrow caret ^ to EVERY caption group that occurs at the same time as

the burned-in text, even if it occurs for only a split second.

ACCURACY

Type what the speaker says.

 Never correct (edit) the speaker's grammar.

 Never paraphrase or substitute words.

 Never insert words not spoken.

 Never rearrange the order of speech.

 Don't correct phonetics unless it distracts from readability.

 Do remove speech disfluency that distracts from readability.

Pay extra attention to the spelling and capitalization of special words.

- Research spellings and terminology you may not be familiar with.

Take time to research the proper spelling and capitalization of important words and proper nouns.

 Research proper spellings and capitalizations for important words. Google proper spelling of

proper nouns (e.g., names, brands, and places) and topic-specific vocab (e.g., Adobe

Premiere Pro).

When you cannot confidently hear or understand a word, use an atmospheric.

 Use an appropriate atmospheric such as "(mumbles)" for a single speaker or "(background

noise drowns out other sounds)". - It's a long drive to (mumbles).

It is NEVER acceptable to use (inaudible), (unintelligible), or ??? in captioning.

SYNCING

Start Time of a Caption Group

 The start time should align with the beginning of the sound. This applies to both

atmospherics and speech.

 While the video plays, press the UP arrow key on your keyboard to align the start of a

caption group precisely when the first word or sound of a caption group is heard.

o Note: When you press the UP arrow key, Dash automatically bumps your timestamp

forward by 1/4 second to compensate for normal lag in human reaction time.