All terms

Data Annotation

Labeling raw data (text, images, audio, video) so it can be used to train or evaluate an AI model.

Data annotation covers a wide range of tasks: tagging objects in images, transcribing audio, marking the sentiment of a sentence, or flagging whether a video contains a specific action. It's usually less specialized than model evaluation work and pays less per hour, but has a lower barrier to entry and higher task volume on most platforms.

Example

Drawing a bounding box around every car in a set of street-scene photos, or labeling each sentence in a support ticket as "angry," "neutral," or "satisfied."

Common questions

It's usually on the lower end of the pay range because the barrier to entry is low, but volume tends to be higher and steadier than specialist evaluation work.

Most platforms provide their own in-browser labeling tool - you rarely need to install anything beyond what the platform gives you.

Related terms