Many social media and photo collection applications offer features to generate a personalized digest of photos or structured text. A digest typically includes highlights of content from a larger collection. Such digest generation and viewing capabilities are not currently available for freeform unstructured conversational interactions, such as those that take place via messaging or chat applications. This disclosure describes techniques to automatically generate a digest of such interactions with permission from the conversing parties. The conversational interactions may contain a mixture of text, images, audio, and video. A user can review and save the digest offline for later personal use, share the digest with the other conversation parties, or post it on social media. The described techniques can be applied to direct one-to-one conversations as well as to multi-party group conversations.

Creative Commons License

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