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.

Menu

This guide takes you through the process of generating and reviewing automated transcripts for audio and video files using the Content Hub platform and Originator tool (for human QC and review). This guide applies to all users responsible for submitting, reviewing, and finalizing automated transcripts.

Once your transcription is created, you can learn more about editing the transcription in Originator Studio here.
Also, to power character attribution for automated transcripts, please make sure to first have the character (or contributor or participant depending on your content type) set up here in the Character Library. 

Navigate this article:

Finding the Automated Transcriptions Tool

Note: this tool is still in beta and, therefore, may not be available in your Content Hub project. If you don’t see the tool available to you, please reach out to your Netflix contact.

  1. Accessing Content Hub:
  2. Navigating to the Tool Library page:
    • Within your project, go to the Tool Library page in the lefthand navigation side panel.
  3. Find the Automated Transcriptions tool:
    • Locate the Automated Transcriptions  tool and then select anywhere on the card to open the tool. This will open the tool within the library.

Once you land in the Automated Transcriptions tool, you will see two main tabs present: the Transcriptions dashboard and the QC Assignments dashboard. Use the table of contents at the top of the guide to skip ahead to learn more about these dashboards.

There is also a blue New Transcription button at the upper right corner of the page. 

Creating a New Transcription

To create a new automated transcription job, select the blue New Transcription button. This will open the New Transcription page. 

Start naming your transcription by entering it into the field that says Enter transcription name. As you enter the name, you will see the folder structure naming change in real time below under Output Formats.

Note: Name your transcription to easily identify it in the dashboard. Transcription names must be unique and cannot be changed after creation. It can only contain letters, numbers, and hyphens.

Once you have entered the transcription name and it matches the naming requirements you will see the Continue and select files button become active and turn blue. Select the Continue and select files button to proceed.

You have two options when selecting files for this transcription job: 

  • Workspace files
    • Locate and select files that have already been uploaded to your Workspace in Content Hub for this project. 
    • For large files, it is recommended to upload them to your Workspace before submitting (instead of using the Upload Files option with the tool).
  • Upload Files
    • Manually upload files from your local machine.

Note: though there are two options, you do not have to choose only one when setting up a new transcription. You can add some files manually and select some files from your Workspace folder -- all in the same submission flow. 

In either case, for the best experience it is recommended that you only submit supported file formats for transcription.  

Supported Formats & Media Input Guidance

  • Audio (Recommended) 
    • File Extensions 
      • .wav
      • .mp3 
        • Note that MP3 files cannot preserve timecode in their metadata, so please follow the {filename}_HHMMSSFF.mp3 format to retain timecode in the transcript.
      • .aac and .m4a are also supported, 
        • Note: needs to follow {filename}_HHMMSSFF.aac format to retain timecode in the transcript
    • Stereo configuration preferred.
    • Max file size:  5 GB
      • Only applicable for the Upload Files option, which uploads through the web browser and not Workspace files, which uploads via the Aspera Connect client.
      • File should not exceed 3 hours in duration
  • Video (Recommended)
    • File extensions 
      • MOV
      • MP4
      • .AVI and .MKV are also supported
        • Note: needs to follow {filename}_HHMMSSFF.aac format to retain timecode in the transcript
    • Codecs & Specs
      • In general, the H.264/H.265 video codecs are preferred, as they typically result in smaller file sizes.
      • Standard constant frame rates such as 23.976, 24, 25, 59.94, and 60. (Variable frame rates not supported)
      • Stereo AAC for embedded audio is recommended.
      • It is recommended for timecode to be embedded in the file (this tool cannot read on-screen timecode)
    • Max file size: 5 GB
      • Only applicable for the Upload Files option, which uploads through the web browser and not Workspace files, which uploads via the Aspera Connect client.
      • File should not exceed 3 hours in duration

Workspace Files

When using the Workspace files tab, you will see the parent level directory of your My Workspace with three columns - Name, Transfer (status), and Size. To expand a folder, simply click the caret icon next to the folder name. Once you have found the file(s) you would like to submit for transcription, select the row of the file and you will see it highlighted in blue. You will also see the files added to Selected Files lower on the page.

Note: more than one file can be selected here by clicking on another row, which will also highlight it in blue. Click the row again to de-select. 

When ready select, Configure Settings to move on to the next step.

Upload Files

If you choose to instead manually upload files from your local machine at the time of creating a submission, select the Upload files tab. From here, you can either Drag and drop files into the drop area or click the file browser by selecting click to browse. 

For Drag and drop, the entire area within the dotted line can be used as a drop zone. Once the file(s) are dropped, you will see them appear below queued for Upload lower on the page.

To find files on your local machine, select  click to browse. The will open up your machine file browser. Navigate your folder structures and select the file you wish to add. Then select Open.

In either case once you have added files, you will see them represented below the drag-and-drop zone area where they will be queued up for upload. 

If you mistakenly added the wrong file, you can select the Reset Files button which looks like an undo arrow. You may also select the X in the upper right. 

Once you have the right files selected and everything looks good, select the blue Upload button to initiate the upload.

Note: it may take a few moments for the upload to complete.

Once the upload is complete, you will see the progress indicate 100% , a green checkmark, and you will see the uploaded files added to Selected Files. To proceed to the next step, select the blue Configure Settings button, which will take you to the Configuration page. 

Configuration Settings

Regardless of how the files were selected, the experience on the Configuration page remains that same. Here you can:

  • Configure Auto-Detect Language feature
  • Assign Human QC for this automated transcription
  • Choose Outputs Formats

By default,  Auto-Detect Language feature will be toggled on.

To assign a human to provide quality control to this automated transcription, click into the box that says the Select additional reviewers under Addition Reviewers section. 

This will reveal a list of eligible reviewers you can select a person to assign. Note: additional human review QC is optional.

Important to note: for a person to appear in the Select Reviewers field, they must have access to the project. Reach out to your Netflix point of contact if you aren’t seeing who you want to assign in the dropdown.

Select the checkbox next to the person’s name and email that you want to assign. The person(s) assigned to QC will receive an email notification once the transcription is completed and it is ready for QC.

More on the QC process here.

When configuring your transcription, here are the available options for Output Formats:

  • TTAL (required) - timed text markup language
  • avidTxt - standard Avid transcription format 
    • Compatible with Avid Media Composer and ScriptSync
  • JSON - structured data format
  • SRT - SubRp subtitle format
    • Compatible with Adobe Premiere and DaVinci Resolve

To choose Output Formats, select the checkbox next to your desired format. (You can select as many as desired but TTAL is a required format for the workflow.)

Once you have finalized your configuration and everything looks good, you can select the blue Start Transcription button.

Tracking Progress

Once you have started the transcription, you will proceed to a Progress page that includes all the Transcription details, a list of the Files that will be output, as well as an overall progress bar.

After a file completes processing, you will see different buttons become active: Workspace, Unreviewed Output Folder, and Originator Studio.

CH_ToolLibrary_AT_18.gif

Selecting the Workspace button will open a new tab and take you to your Workspace where a new folder structure has been created for the transcription you submitted. The name of the folder structure will match the name you entered for the submission (and will also reside in a more generic Automated Transcription folder). 

Within the specific submission folder are two subfolders: input and output. Selecting Workspace will specifically take you to the output subfolder. This provides access to the output files that were just generated from your transcription. From here, they can be downloaded or shared.

Selecting the Unreviewed Output Folder button will open a new tab and take you specifically to the unreviewed subfolder within the output subfolder in your Workspace.

From here, these files can be shared for QC outside of the Originator Studio platform, for example.

Selecting the Originator Studio button will take you out of Content Hub and into the Originator Studio platform, where the human QC will take place.

Note: additional permissions are required to have access to Originator Studio to perform the QC.

Transcriptions Dashboard

From the Transcription Details page, you can return to the Transcription Dashboard at any time with ← Back to Dashboard button.

Also, when you enter the Automated Transcription tool, it will default to the Transcriptions dashboard tab. 

CH_ToolLibrary_AT_23.gif

Once you enter the dashboard, you will see four sub-tabs that act as quick filters: 

  • All
    • Shows a table of all automated transcription jobs that have been submitted on this show by anyone. 
  • Active
    • Shows a table of all active or in-progress automated transcription jobs that have been submitted on this show by anyone. 
  • Completed
    • Shows a table of all complete automated transcription jobs that have been submitted on this show by anyone.
  • My Submissions
    • Shows a table of all automated transcription jobs that have been submitted on this show by you - regardless of Status.

Navigating the Transcriptions Dashboard

Near the upper right of the dashboard, you can toggle between Table View and Card View. 

Regardless of what view you are in, there is a search bar at the top where you can Search for a transcription to quickly find what you are looking for. This searches against the name of the automated transcription submissions only. 

Table View

When in Table View, each transcription submission is represented by a row on the table with four columns:

  • Transcription Name
    • The custom name given to the automated transcription job at the time of submission.
  • Status
    • Current status of the automated transcription job - can be In-Progress, Completed, or Cancelled.
  • Submitted
    • The email of the Content Hub user who submitted the automated transcription job.
  • Date
    • The date of the submission.
    • Pro Tip: hover over the date to see Pipeline Start Date and Pipeline Finish Date, which also includes time of day so you can see how long the job took start to finish. 

At the bottom of the table, you can also customize the table itself by choosing how many Rows per page. In the case that there are multiple pages of results, you can toggle between pages using the right (>) and left (<) caret icons in the bottom left of the table. 

CH_ToolLibrary_AT_28.gif

Card View

When in Card View, each transcription submission is represented by a card. In Card View, you get all the same information as the Table View but with the added benefit of seeing the output formats.

Transcription Completed & Ready to Download - Email Notification

When a transcription has completed successfully, an email will be sent for all transcription files from that submission. If a transcription is submitted with 3 files, the email will list all 3 of those files. 

The recipients of this email are the person who submitted the transcription as well as anyone who was assigned as a reviewer to alert them that the transcription(s) are now ready to review.

The email sent to the reviewer will let them know that the transcription was successfully generated, but has not yet been reviewed and is ready for download or QC.

Transcription Details Page 

To see more Transcription Details about a specific automated transcription job, click to select any place on the row or click to select anywhere on a card. In either case, this will open the Transcription Details page.

From the Transcription Details page, you see the following information:

Across the top, there is at a glance a reference of Total submitted files, the Status, the number of Selected output formats.

You can also see the following details:

  • Progress bar
  • Submission name
  • Status
  • Created Date
  • Selected formats
  • Human QC enabled (Yes / No)
  • Reviewers
  • Total files

At any time, you can select the ← Back to Dashboard button to return to the Transcriptions dashboard.

QC Assignments Dashboard

To enter the QC Assignments dashboard, select QC Assignments. 

Once you enter the dashboard, you will see two sub-tabs that act as quick filters: 

  • Pending
    • Shows a card view that filters down to any transcriptions that have been assigned to you for this title and have not yet been completed and are still in-progress. The same buttons from this step are present within the card.
  • Completed
    • Shows a card view that filters down to any transcriptions that have been assigned to you for this title and have been completed. The same buttons from this step are present within the card.

Transcription Editing & Quality Control in Originator Studio

If you have been assigned to QC the transcription (where you can also make edits), follow the Originator Studio button link. Please also refer to this guide for more details Transcription Editing & QC in Originator Studio. 

Transcription QC Completed - Email Notifications

When quality control has been completed and a transcribed file has been approved in Originator Studio, an email will be sent automatically. In this case, multiple QC completed emails could be sent for a single submission - one for each file that was submitted. For example, if 3 files were submitted, up to 3 emails could be sent once each file has completed QC.

Each email will only be sent to both the person who submitted the transcription.

 

 

 

Was this article helpful?
0 out of 0 found this helpful