A Step-by-step Guide to ‘AudiAnnotate’


Annotation is an essential part of all disciplines, and the Digital Humanities are no exception, but we (or at least I) tend to associate annotation with textual works. In this tutorial I will outline how to utilize AudiAnnotate to create annotated video or audio. The web application itself is a relatively straightforward way for scholars and other observers to translate their analyzes of audio/video into annotations.

This is an invaluable tool for documenting our understandings of auditory digital humanities. This provides scholars with an effective way to annotate their observations while making the material more engaging to an outside audience through their added interpretation.

Walk through:

  • Step 1 : Signing in!

Since this project is hosted on GitHub, it will require you to sign in with your account (so create one if you have not already). This link will take you to the below page where you can sign in (near the top right). A pop up may appear asking you to authorize access to your GitHub account; you can just click authorize.

Apart from just signing in, on the right hand side there is a column of example of annotated audio to explore. Note that audio can be both pure audio or videos, so this tool can create annotations for both!

  • Step 2 : Creating a new project

You will then be brought to the following page. Simply hit ‘New Project‘ to begin the process of making your own annotated audio.

  • Step 3 : Titling your project

You will then be presented with the below page, which prompts you to fill out three fields describing your project. There is a certain amount of jargon that may or may not make sense to you depending on your level of familiarity with Git, but the TLDR is your project will be saved in the cloud (or your GitHub), and you are just filling out the necessary metadata of your project.

Your title will be whatever name suits your project (ie. Carleton Inaugural speech), the description (ie. Alison Byerly is installed as Carleton’s 12th president), and the project slug which will name the ‘folder’ storing your project (ie. Carleton-Inaugural-speech). Note that no spaces are allowed in your project slug.

  • Step 4 : Create a new ‘Item Manifest’

Once your new project has been created, you will be presented with the following screen. For our purposes (creating an entirely new annotation for a video or audio), we will select ‘Create Item Manifest‘.

  • Step 5 : Metadata

Now that we are creating a new Item Manifest, there are a few fields of metadata we need to supply (only those labeled with an asterisk are required). Labeling your manifest is required, and I recommend relating it to your project. The audio file URL is perhaps the trickier part. You cannot simply paste in the link of a youtube video; it needs the direct link. A direct link should open the item in a web browser as a standalone object, begin as a web address, and end with the file format, e.g. .mp3, .mp4, .wav, etc. There are a few ways to get the direct link from a video. The easiest was might be using some kind of third-party software that supplies you with a video download, like y2mate. From there you need to copy the file path of the file (you can see mine below). After that, all you need to do is specify the length of the video.

After selecting ‘Save’, you should be presented with the following screen (I was bombarded with scary looking errors, but when I refreshed everything was fine).

  • Step 6 : Creating Annotations using a spreadsheet

We will be moving outside of AudiAnnotate for the next step. There are a few ways to create annotations for your video, but I think a spreadsheet is the simplest and easiest way. There are four columns:

  • The first column should contain the beginning time (either in total seconds or hour:min format) of the annotations.
  • The second column should contain the end time of the annotation if necessary (you can either create a point or range annotation where a range has a beginning and end while a point has one time).
  • The third column should contain the contents of your annotation.
  • The fourth column should contain the name of the layer. The layer column is where you can differentiate types of annotations. Write your layer names as you would like them to appear in AudiAnnotate (e.g., “Speaker”, “Audience”).

Once you are done adding annotations, save your spreadsheet as a csv (comma separated values).

  • Step 7 : Adding your annotations to your project

Navigate back to your projects page in AudiAnnotate and select ‘Edit’ on the appropriate project.

Now click on the title of the Item Manifest we created in steps 4 & 5. Notice where my cursor is hovering.

After selecting that md file, you will be presented by the following screen. Here is where we will add the annotations we created in step 6. Click the ‘Choose File’ button where my cursor is hovering, and select the csv file you created. Once you have the appropriate file, select the ‘Add’ button directly below.

You will then have to define which columns specify the start time, end time, annotation, and layers. They should correspond to the letters at the top of each column in your csv. Once you have filled out the appropriate fields, select the ‘Process’ button.

Once you see the green ‘Ready’ alert in the top right of the screen, you can visit the project site to see your annotations! It is highlighted below.

  • Step 8 : Exporting from GitHub

For the final step, we will export our finished project from our GitHub account. Naviagte to your GitHub account and select ‘Repositories’ to view the repository that is housing your project. To export, go to the repository page and look for the green ‘Code’ button. Select ‘Download ZIP’. The page should look like the following:

And you should find your project in your Downloads folder as a zip file!

Other resources:

Here is an article reviewing AudiAnnotate as a tool for annotating audio files. And here are video guides posted by its creators.

-Liam Keane 2/28

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.