Search

US-20260128921-A1 - AUTOMATED GENERATION OF MEETING TAPESTRIES

US20260128921A1US 20260128921 A1US20260128921 A1US 20260128921A1US-20260128921-A1

Abstract

The disclosed technology teaches a method for automatically generating a graphical summary of a meeting that provides a reflection of a group's conversation with textual and graphical elements in a tapestry. The method comprises a combination of transcript analysis using natural language processing or other machine learning models and automatic selection of graphic symbols detected as relevant to the transcript by statistical analyses. The method further includes the ability to interact with a generated tapestry for additional personalization.

Inventors

  • Eileen Marie Clegg
  • Benjamin Mandeberg
  • Ramana Rao

Assignees

  • vTapestry, Inc.

Dates

Publication Date
20260507
Application Date
20251230

Claims (19)

  1. 1 . A computer-implemented method for generating a graphical representation of a human conversation comprising textual and graphical elements in a tapestry, the method comprising: receiving conversational content comprising text from participants in a conversation; processing the conversational content with a language understanding model to extract textual elements; determining using the language understanding model textual features of the textual elements, including one or more of syntactic features, semantic features, sentiments or themes; applying a large language model (LLM) or rule to the textual features and corresponding text and symbols, including: (a) determining spatial placement of the text and symbols using the LLM or a placement mapping rule to defining positional regions of the tapestry; (b) determining visual mapping of the textual features to the text and symbols using the LLM or visual mapping rules to modify perceptual traits of the text and symbols, including typography, font, font size, or font color and aesthetic design, size, or color of symbols; and (c) determining symbol-selection using the LLM or symbol selection rule to select which graphical, image, rendering, symbolic depiction, animation, camera- or digitally-generated graphic is chosen to represent a textual element; representing the textual elements as text and symbols that are spatially arranged and visually modified on the tapestry.
  2. 2 . The computer-implemented method of claim 1 , wherein the conversational content is from an online remote meeting.
  3. 3 . The computer-implemented method of claim 1 , wherein the conversational content is parsed for textual elements, time length of speech, time length between participants speaking, sentiment scoring, and annotation data.
  4. 4 . The computer-implemented method of claim 1 , further including accessing a tapestry template that includes rules for placement mapping, visual mapping and symbol-selection and using the tapestry template and the included rules.
  5. 5 . The computer-implemented method of claim 4 , wherein the tapestry template comprises textual element slots and graphical element slots, wherein an arrangement of content into the slots follows the included rules.
  6. 6 . The computer-implemented method of claim 5 , wherein the included rules for the arrangement of slots include rules dictating a number of slots within the template, a size of the slots, or a location of slots.
  7. 7 . The computer-implemented method of claim 1 , further including accessing and applying a filter that specifies inclusion of one or more symbols.
  8. 8 . The computer-implemented method of claim 1 , further including accessing and applying a prioritization schema for graphical element placement within a tapestry.
  9. 9 . The computer-implemented method of claim 1 , wherein the LLM processes a frequency of a particular textual element.
  10. 10 . The computer-implemented method of claim 1 , wherein the LLM processes a set of weighted importances corresponding to textual elements.
  11. 11 . The computer-implemented method of claim 1 , wherein words processed within the conversational content by the language understanding model are classified within a particular lemma.
  12. 12 . The computer-implemented method of claim 1 , wherein a relationship between textual element data structures and graphical element data structures is many-to-one, such that a plurality of textual elements may relate to the same graphical element.
  13. 13 . The computer-implemented method of claim 4 , wherein the tapestry template comprises a plurality of rules, and wherein at least one of the plurality of rules requires that a particular slot can only be filled with a graphical element or a textual element.
  14. 14 . The computer-implemented method of claim 13 , wherein a rule within the plurality of rules states that one or more textual or graphical elements are prioritized for a particular slot.
  15. 15 . A tangible non-transitory computer-readable storage media, including program instructions loaded into memory that, when executed on processors, cause the processors to implement a method for generating a graphical representation of a human conversation comprising textual and graphical elements in a tapestry, the method including: processing, as input, a transcript from a meeting with a natural language processing (NLP) model to extract, as output, textual elements including lemmas, words, phrases, and sentences; selecting textual elements among extracted textual elements based on statistical analysis of the transcript; selecting graphical elements from a symbol library based on the statistical analyses of the selected textual elements; filling slots within a tapestry template with the selected textual and graphical elements; and generating a tapestry including the selected textual and graphical elements within filled slots.
  16. 16 . The non-transitory computer-readable storage media of claim 15 , wherein the tapestry template comprises textual element slots and graphical element slots, wherein an arrangement of content into the slots follows one or more pre-defined rules.
  17. 17 . A system for automatically generating a graphical summary of a meeting that provides a reflection of a group's conversation with textual and graphical elements in a tapestry, the system including a processor, memory coupled to the processor and program instructions from the non-transitory computer-readable storage media of claim 15 loaded into the memory.
  18. 18 . The system of claim 17 , wherein the tapestry template comprises a plurality of rules, and wherein a rule within the plurality of rules states that a particular slot can only be filled with a graphical element or a textual element.
  19. 19 . The system of claim 18 , wherein a rule within the plurality of rules states that one or more textual or graphical elements are prioritized for a particular slot.

Description

PRIORITY This application is a continuation of U.S. patent application Ser. No. 18/377,260, titled “Automated Generation Of Meeting Tapestries,” filed 5 Oct. 2023 issued 30 Dec. 2025 as US 12,513,019 (Atty docket no. VTAP 1000-2), which claims priority to and the benefit of U.S. Provisional Application No. 63/416,914 titled “Automated Generation of Meeting Tapestries,” filed 17 Oct. 2022 (Atty Docket No. VTAP 1000-1.) RELATED CASES This application is related to the following contemporaneously filed applications: U.S. patent application Ser. No. 18/377,263, titled “Artificial Intelligence Model Generation of Meeting Tapestries,” filed 5 Oct. 2023 issued 9 Dec. 2025 as U.S. Pat. No. 12,494,933 (Atty docket no. VTAP 1000-3); and U.S. patent application Ser. No. 18/377,270 titled “Interactive Generation of Meeting Tapestries,” filed 5 Oct. 2023 issued 29-Oct-2024 as U.S. Pat. No. 12,132,580 (Atty docket no. VTAP 1001-1); The related applications are incorporated by reference herein for all purposes. FIELD OF THE TECHNOLOGY DISCLOSED The technology disclosed generally relates to generative artificial intelligence models that begin with language understanding and generation tasks such as those within the domain of natural language processing (NLP), and more specifically to the application of artificial intelligence to the automatic generation of a graphical summary of a transcript that provides a reflection of a conversation with images, key words, sentences, and phrases in a digitally-rendered tapestry. BACKGROUND The subject matter discussed in this section should not be assumed to be prior art merely as a result of its mention in this section. Similarly, a problem mentioned in this section or associated with the subject matter provided as background should not be assumed to have been previously recognized in the prior art. The subject matter in this section merely represents different approaches, which in and of themselves can also correspond to implementations of the claimed technology. Online meetings have become increasingly common, particularly due to the COVID-19 pandemic beginning in early 2020, and the trend is likely to continue. Disengagement during online meetings has become a rising issue for organizations. Media Naturalness Theory assumes the human brain evolved for face-to-face communication. According to Media Naturalness Theory, online meetings create an unnatural situation of prolonged eye contact with others and information overload while missing critical aspects of trust-building human social experience, such as body language and synchronicity that allow for spontaneous exchange. In addition, the discontinuous communication style inherent to remote teamwork often results in a lack of tangible, memorable takeaways from meetings. The practice of graphic recording addresses this dilemma via an art-based solution targeted at increasing engagement and creativity. The practice has been credited with trust-building because it provides an accessible “big picture” summary of a meeting that shows participants' ideas in juxtaposition with each other, as well as a group memory after the meeting. Studies in the field of neuroaesthetics (the intersection of neurology and the arts) show that art releases neurotransmitters associated with positive emotions. Specifically, landscape and nature scenes have been identified as having a positive effect on the parasympathetic nervous system, lowering stress. Online meeting tools for collaboration typically focus on text alone, such as outline summaries. In contrast, the use of a graphical summary produces art that enhances well-being as a framework for content, supported by neuroaesthetics research. In addition to building collective memory, art also promotes health and well-being for team members. An opportunity arises for an art-based online meeting tool that summarizes the conversation, humanizes the meeting process with nonverbal communication, captures the key themes of a meeting, and invites engagement and creative exploration from team members. Accordingly, an opportunity also arises to leverage algorithmic layout methods that combine artificial intelligence models such as NLP with graphic reporting to automatically generate a graphical summary of a meeting that provides a reflection of a group's conversation with key phrases in a tapestry. BRIEF DESCRIPTION OF THE DRAWINGS The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee. The color drawings also may be available in PAIR via the Supplemental Content tab. In the drawings, like reference characters generally refer to like parts throughout the different views. Also, the drawings are not necessarily to scale, with an emphasis instead generally being placed upon illustrating the principles of the technology disclosed. In the foll