Welcome to the Netflix Partner Help Center. Have a question or need help with an issue? Send us a ticket and we'll help you to a resolution.

A Pivot Language Dialogue List (PLDL) will be commissioned for any content for which the source language is not English. Localization partners servicing non-English content will receive a PLDL as part of the reference materials provided by Netflix. The PLDL will encompass a time-coded transcription of the original dialogue alongside its translation into English. The English translation should remain true to the original creative intent and will serve as the pivot language for consistent translations across all dubbed languages.

 

Scope of Work

The creation of a Pivot Language Dialogue List involves two steps:

  • Original dialogue transcription by a native speaker of the source language.
  • An English translation by a native English speaker with professional knowledge of the source language of the video. If the translator is not a native English speaker, then a proofreading pass by a native English speaker is required to ensure native fluency of the translation.
    • The English translation should:
      • Not take into consideration timing, lip sync or reading speed concerns.
      • Be the most natural, comprehensive and idiomatic rendition of the original text, preserving the tone and nuances creatively intended.

 

The PLDL service will include up two deliveries / completions in the Script Authoring Tool: 

  • A PLDL created to the locked cut; and
  • A PLDL created to final media, alongside a change log of Locked to Final changes pertaining to dialogue and on-screen text.
    • For workflows in which a Netflix QC is required, QC notes should be implemented in the final delivery of the PLDL.
      • Unless specifically requested there is no need to create an updated locked cut version (to implement QC notes) or conform to an intermediary proxy (prior to the final proxy delivery). Any additional versioning is subject to conform rate and will be at the Netflix representative’s discretion.

Script partners will be allocated work via Grand Bazaar and Backlot, then asked to assign their resources in Netflix’s Content Localization Portal, the central location to manage the authoring process.

Once authoring has completed, the author will submit in the tool, which will auto-update the source management request - Pivot Language Dialogue List.

In addition to PLDL creation, the PLDL partner is responsible for the below KNP entries:

  • Key source terms and phrases in the original language
  • English translations for all original language terms
  • FYI - the KNP is accessible from the Authoring application via the reference materials tab 
    • Screen_Shot_2020-08-06_at_1.34.10_PM.png

Reference Materials

Shooting script and KNP tool are all available to access via the same tab. When there is a shooting script available, Authoring tool will import characters in the shooting script by clicking 'Import Characters'. 

  • Screen_Shot_2020-09-28_at_1.21.40_PM.png

Style Guide

The tool is set up to output a format that best suits our partners and internal stakeholders. Therefore, the script partner and authors should not have to worry about change logs, project info, or format specific to previous expectations.

 

Dialogue List

  • Authoring in the tool consists of transcription editorial and various notes tagging. Each dialogue event is a contained unit and organized linearly in the dialogue editor section. Each event will require some level of editorial: In-Timecode, Out-Timecode, Source, Dialogue transcription. 
    • Dialogue Editorial
      • Timecode should be frame-accurate to Netflix burn-in timecode. Timecode is automatically pre-laid for each dialogue event based on automatic speech recognition.
      • Source: Speaker names should be in all caps and include consistent, correct spellings for speakers (according to source material such as shooting script, credits or on-screen text). Unless multiple characters are saying the same dialogue at the same time, only include one speaker/source per event (exception: if dual speakers are included in the burned-in forced subtitle). For multiple people speaking in unison, include names separated by a forward slash in source event.
      •  Examples:
        • Screen_Shot_2020-08-06_at_1.38.27_PM.png
          • All CAPS per Character
        • Screen_Shot_2020-08-06_at_1.38.56_PM.png
          • Multiple speakers in unison style
      • Dialogue should contain: 
        • A completely accurate transcription of all dialogue and stutters/syllables.
        • Descriptions of mouth sounds in brackets (e.g.: [sighs], [laughs], [screaming], etc.) in the dialogue event, before dialogue text begins. 
        • Descriptors that do not apply to specific dialogue or describe mouth sounds [e.g.: (long pause), (fades out to shot change), etc] in parentheses.
        • Best Practice: Utilize finalize dialogue stems to transcribe difficult to hear dialogue and individual lines within larger walla beds. Reach out to Netflix to see if available.
      • Translation should contain the English translation of all dialogue as well as on-screen text.
      • If there are burned-in English production subtitles to the original dialogue in the locked proxy, those do not need to be transcribed into the English translation.
    • Annotations
      • The PLDL should include annotations for any instance of challenging idiomatic expressions. These annotations are required to ensure downstream localization remains as close to the original creative intent as possible.
      • When creating a note, the user is prompted to select the type of note then add any additional text to the note. 
      • Pay special attention to the inclusion of local idioms, slang, cultural references, double entendres, unclear references, the pronunciation of foreign names, etc. 
      • Example: 
      • Screen_Shot_2020-08-06_at_1.42.31_PM.png
      • Be mindful of idioms that are engraved in culture and day-to-day life. In both source language and English. Keep in mind that direct translation of idioms can change the meaning of the entire dialogue.  
    • Types of Annotations
      • Context annotations: please include annotations for words and phrases - to be filled out in “Annotation Notes” box
      • On-Screen Text - Burned in subs - should contain transcriptions of production-approved burned-in subtitles (for foreign dialogue, hard-to-hear dialogue, or sign language) as they appear in that cut of picture. Please include the character name in the source box in the authoring tool and transcription of On-screen text in the main dialogue event box. 
        • Content/timing of subtitles should match the most recent cut and/or document provided by production. Line and subtitle breaks should be honored as they appear in documents.
        • Please split continuous dialogue up into rows according to corresponding subtitles if applicable. If the timing of dialogue/subs are not the same, time according to dialogue. 
        • If burned-in subtitles contain content translating foreign dialogue, the dialogue must contain a description (e.g. [speaking spanish]) or transcription of the foreign dialogue in the same row. 
        • For hard-to-hear dialogue with burned-in subtitles, the dialogue event should still contain a transcription that will often match the content in the burned-in subtitle column.
        • For sign language with burned-in subtitles, the dialogue event should be left blank as no dialogue is being spoken.
        • Note: Final picture will often be clean of subtitles, but this content should stay in the final Pivot Language Dialogue List and be timed to match the final version of picture as it would occur had it been left in.
        • Example
          • Screen_Shot_2020-08-06_at_1.42.35_PM.png
      • On-Screen Text - other types - should include all prominent, pertinent on-screen text in the annotations editor. Please be accurate with your description of the event via the dropdown options and free form notes section for further context. Please use the verbiage NONE in the source box in the authoring tool - this allows for the tag type to direct the text to the proper column in the xls export. 
        • On-Screen Text - Graphics & Inserts
        • On-Screen Text - Principle Photography
        • On-Screen Text - Episodic Title
        • On-Screen Text - Netflix Credits
        • On-Screen Text - Main Title
        • On-Screen Text - Netflix Credits
        • Place transcribed & translated in the main dialogue box
        • Leave annotation notes section for annotations
        • Screen_Shot_2020-08-06_at_1.42.39_PM.png
      • Foreign dialogue treatment:
        • Foreign words including short phrases:
          • Transcribe in the original language and use the tag ‘Foreign Dialog’
      • Longer Passages
        • Indicate foreign dialogue language 
          • use format [speaks German]
      • Transcribe foreign language
        • If foreign dialogue is creatively intended to be understood (i.e. subtitled in the original version), translate into English (using burned-in subtitles in locked cut as reference) 
      • *If any mistranslations are noted in burned-in subtitles, please flag to Netflix before updating in the dialogue list.
    • Songs & Ditties should contain transcriptions of vocals, whether licensed, un-licensed, original IP, or ditties. 
      • Screen_Shot_2020-08-06_at_1.42.41_PM.png
    • Archival Footage - TV news footage, court footage, previously released movie/TV content, etc.; i.e., anything filmed outside of the film/series itself should be transcribed and tagged as archival footage.
  • Off-screen dialogue treatment:
    • When a character speaks off-screen (still in-scene but not on-camera), please indicate by putting (OFF) at the end of the character’s name in the dialogue event.
    • If the character’s dialogue section alternates between on- and off-screen please indicate the order that it occurs, e.g.: (OFF/ON) for a character’s dialogue that starts off-screen and continues on-screen, (ON/OFF) for starting on-screen and going off-screen. No need to include more than two cues.
    • If a character is providing voice-over (not in-scene at all), please start new time-coded dialogue entry for when voice-over begins and include (VO) at the end of the character’s name in the event.

Service Level Agreement (SLA)

  • To be considered on-time, the partner must deliver the initial translation within the agreed SLA of receipt of video. Requests for delivery in less than the specific SLAs according to runtime will be subject to rush fee.
  • Specific SLAs are as follows with runtimes rounding up to next 30-minute interval:
    • 30 min TRT - 5 business days
    • 60 min TRT - 6 business days
    • 90 min TRT - 7 business days
    • 120 min TRT - 8 business days
  • Versioning and conforms should be delivered within 48 hours.
  • Failure to meet requirements and style guide specifications will result in redelivery requests at partner cost that will impact metrics.

Change List

  • Auto logged in the xls export (Available as an export in-tool, or downloadable via backlot)
    • This list will include changes/updates from the previous delivery of the Dialogue List.
    • Granular details about updated dialogue/on-screen text and summaries of global or broad changes (e.g.: eight-second offset at beginning of program, scene added at 01:22:30, etc) should be noted on the Annotations column.
    • Please refrain from using formatting (bold, underline or strikethrough) to indicate changes, as these might be lost when importing documents to different tools.
Was this article helpful?
2 out of 2 found this helpful
Powered by Zendesk