17967714. TRANSCRIPT QUESTION SEARCH FOR TEXT-BASED VIDEO EDITING simplified abstract (Adobe Inc.)

TRANSCRIPT QUESTION SEARCH FOR TEXT-BASED VIDEO EDITING

Organization Name

Adobe Inc.

Inventor(s)

Lubomira Assenova Dontcheva of Seattle WA (US)

Anh Lan Truong of San Carlos CA (US)

Hanieh Deilamsalehy of Seattle WA (US)

Kim Pascal Pimmel of Seattle WA (US)

Aseem Omprakash Agarwala of Seattle WA (US)

Dingzeyu Li of Seattle WA (US)

Joel Richard Brandt of Venice CA (US)

Joy Oakyung Kim of Sunnyvale CA (US)

TRANSCRIPT QUESTION SEARCH FOR TEXT-BASED VIDEO EDITING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17967714 titled 'TRANSCRIPT QUESTION SEARCH FOR TEXT-BASED VIDEO EDITING

The present invention provides a system, method, and computer storage media for searching for meaningful questions in a video by transcribing the audio track and identifying sentences that end with a question mark.

• Transcribe audio track from video
• Parse transcript to identify sentences ending with a question mark
• Filter out certain types of questions (e.g., short questions, logistical questions, rhetorical questions)
• Identify questions in response to a command
• Present search result tiles representing video segments of the questions
• Navigate to corresponding portion of the transcript by selecting a search result tile

• • Potential Applications:**

- Video content creation - Educational platforms - Content recommendation systems

• • Problems Solved:**

- Efficiently finding meaningful questions in video content - Enhancing user engagement with video transcripts

• • Benefits:**

- Improved search functionality for video content - Enhanced user experience with video transcripts

• • Commercial Applications:**

- Video streaming platforms - Educational technology companies - Content recommendation services

• • Prior Art:**

There are existing systems for transcribing audio and searching for specific content within videos, but the specific focus on identifying and filtering questions in video transcripts is a novel aspect of this invention.

• • Frequently Updated Research:**

Research on natural language processing and video content analysis may provide insights into improving the efficiency and accuracy of question search algorithms.

• • Questions about the Technology:**

1. How does this technology differentiate between different types of questions in a video transcript? 2. What are the potential challenges in accurately identifying and filtering out questions in video content?

Original Abstract Submitted

Embodiments of the present invention provide systems, methods, and computer storage media for a question search for meaningful questions that appear in a video. In an example embodiment, an audio track from a video is transcribed, and the transcript is parsed to identify sentences that end with a question mark. Depending on the embodiment, one or more types of questions are filtered out, such as short questions less than a designated length or duration, logistical questions, and/or rhetorical questions. As such, in response to a command to perform a question search, the questions are identified, and search result tiles representing video segments of the questions are presented. Selecting (e.g., clicking or tapping on) a search result tile navigates a transcript interface to a corresponding portion of the transcript.