A manager wants a concise version of a long incident report that preserves its central points. Which capability fits?

1. Image segmentation
2. Text summarization
3. Speech synthesis only
4. Named entity recognition only