Documentation

Learn how to use the Multimodal Annotation Tool effectively

Getting Started

Set up your annotation schema (click "Set Up Schema" button)
Upload your media file (audio or video) and transcript CSV file (supports drag and drop)
Start annotating by clicking on utterances and filling in the annotation fields
Export your annotated data when finished

Column	Type	Description
`turn_id`	integer	Unique identifier for each utterance
`speaker`	string	Speaker identifier (e.g., "S1", "S2")
`start`	float	Start time in seconds
`end`	float	End time in seconds
`utterance`	string	The spoken text

turn_id,speaker,start,end,utterance
1,S1,0.0,2.5,"Hello, how are you?"
2,S2,3.0,5.2,"I'm doing well, thanks!"
3,S1,6.1,8.9,"That's great to hear."

MP3 and MP4 formats are highly compressed and provide fast processing with smaller file sizes.

Optimal performance: Keep files under 500 MB
Maximum recommended: 1 GB
Warning: Files over 1 GB may cause slow processing or browser performance issues as all data is processed in-browser memory

Keep CSV files under 10 MB for best performance. Large transcripts with thousands of rows may cause UI lag.

Play / PauseSpace

Navigate annotationsTab

Define custom annotation columns with your own labels. Set up your schema before loading data for automatic column matching.

The transcript automatically highlights the current utterance during playback. Click any utterance to jump directly to that point in the media.

Simply drag media and CSV files directly onto the window for quick loading.

Export your annotated data as a CSV file, preserving all original columns plus your annotations.

Adjust the layout by dragging panel dividers to customize your workspace.

Click the "Save Annotations" button to export your annotated data. The exported CSV will include:

The exported file will be named original_filename_annotated.csv