Picture Book Annotation

How to code the Picture Book audio recordings in ELAN.

Open ELAN

  1. If you do not have ELAN on your device, follow these instructions for download on the ELAN website.

    1. Please make sure to download AVFX version if you are using MacOS High Sierra and above!

Import to ELAN

  1. Open ELAN and create a new file (command/ctrl + n).

  2. Add the "Media" file (the audio file).

    1. If you do not see the audio file, you must convert your audio file using the Audacity Scripting Page.

  3. Add the "Template" file, called PB_template.etf.

    1. You can find it here on the Psych Drive: /Volumes/psych/BergelsonLab/Subject_Files/PBS/ELAN Template

    2. You can also find it online here.

  4. Select OK.

Annotate in ELAN

You can customize keyboard shortcuts to your preference! Go to Edit>Preferences>Edit Shortcuts>Annotation Mode.

Listen to the audio file and pause when an object word is uttered. Annotate all audio recordings for the participant.

The Script fro the book is here.

  1. [Click+drag] Highlight the approximate area of the target word utterance from the book.

    1. Example: "Look at the dax!"

  2. [Shift+click on either side of the area] Adjust the start and end of the area to be more precise.

  3. [Alt+n] Create a new annotation.

  4. [Double+click] Type into the 5 tiers. See the Coding Notes section below for guidance.

  5. Annotate all audio recordings for the participant. Save the .eaf files with the following name format: [SubjectID]_PB_[Recording#]_coder[Initials].

    1. Example: 08_14_PB_01_coderSY.eaf

Coding Notes

The Tiers are as follows:

FA1: Female Adult 1 - This will most commonly be used. If there is a second parent, you can add in MA1 for Male Adult 1, FA2 and so on

CHI: For Target Child Utterances

xds@CHI: Indicate who the child is talking to "A" for Adult.

mwu@CHI: How long is the utterance

xds@FA1: when the are speaking to only the target child, please annotate with "T". When the parent is talking to another child or to multiple children, both including or not including the target child, please annotate with "C".

txt@FA1: differentiate if the speech is directly from the book or if it is part of a conversation with the child. Type B for book and C for conversation.

Additional conventions:

[!= whispers].

&=laughs

&= gasps

Onomatopoeia: When there is onomatopoeia in the transcription, follow with @o.

meow@o said the cat.

splash@o!

Other notes:

  • Code all words.

  • Be vigilant about utterance type! Most of them will be "B" for Book Reading, but some of them will not be.

  • How do I make comments?

    • Double-click on the "comments" tier, and begin the comment with "%com:". For example, "%com: blah blah blah".

  • If the word "apple" is said in reference to shang...

    • Include the comment "Semantic neighbor."

  • How do I deal with mispronunciations?

    • In ELAN, annotate words the way that they are pronounced. (If banoona is read as "boona," type in "boona.")

    • For basic levels, if the word is mispronounced consistently throughout the study, the basic level should be the mispronunciation ("boona"). If the word is only mispronounced once or twice, the basic level should be the correct word ("banoona").

Export Annotated Files to CSV

  1. Annotate all audio recordings for the participant.

  2. In ELAN, go to File > Export Multiple Files As > Tab-delimited Text... > New Domain...

  3. Select the folder that contains all the annotated recordings for the participant, and then select OK.

  4. Save with the following name format: [SubjectID]_PB_basiclevel_coder[Initials].

    1. Example: 08_14_basiclevel_coderSY

  5. Select all the tiers to include.

  6. Check the following boxes...

    1. Separate column for each tier.

    2. Include file name column.

    3. Include file path column.

    4. Begin Time, End Time, and Duration.

    5. ss.msec.

  7. The settings should look something like this...

  8. Select OK. Navigate to where you would like to save the file, and then save the .txt file with the following name format: [SubjectID]_PB_basiclevel_coder[Initials].

    1. Example: 08_14_basiclevel_coderSY

  9. Right-click on the .txt file. Select Open with > Microsoft Excel.

  10. If necessary, rearrange the columns so that they are in the following order...

    1. Begin Time - ss.msec

    2. End Time - ss.msec

    3. Duration - ss.msec

    4. word

    5. utterance

    6. presence

    7. speaker

    8. comments

    9. File

    10. File Path

  11. Add a new column to the left of the "comments" column, named "basic_level". Fill in this column according to the Basic Level Guidelines.

  12. Save your file as a CSV (Comma delimited) in the subject folder.

PreviousPicture BookNextYours/My

Last updated