How to apply this guide
Open the selected image while dictating. Describe its subject and action in one or two sentences.
Start with
Try “Two runners crossing a wet bridge at dawn.” Give the image a meaningful label and keep location details only if needed.
Check before you finish
You can dictate a batch one image at a time, saying the file label before each caption. Review which label belongs to which image before editing the text. Never assume the model can see the image just because it is displayed beside the recorder.
Evidence to keep: Keep the target field open and verify that the complete reviewed result reached the intended destination. This page’s key limit is: Open the selected image while dictating. Describe its subject and action in one or two sentences.
Make a short draft
Try “Two runners crossing a wet bridge at dawn.” Give the image a meaningful label and keep location details only if needed.
Check the destination
Check whether each caption describes the right file. Write separately for alt text when a visual detail is needed for accessibility.
What to watch for
You can dictate a batch one image at a time, saying the file label before each caption. Review which label belongs to which image before editing the text. Never assume the model can see the image just because it is displayed beside the recorder.




