This disclosure describes techniques for automatic augmentation and/or generation of message content. Per techniques of this disclosure, message content is visually augmented based on content recognition and/or user indication. The visual augmentation includes emphasis of specific message content and/or addition of graphical overlay elements based on the message content or user selection. The visual augmentation is portable across platforms and devices. With user permission and express consent, trained machine learning models are used to detect sentiment being expressed in a message and/or key reference points included in the message. Augmentations are suggested based on the detected sentiment or key reference points. Textual cues provided by the user are utilized to surface entities that are related to the textual cues. Based on the entities, user intent and sentiment, a suitable graphical overlay is provided. An augmented message with suggested overlays is displayed to the user, ready for sending or further edits.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.