# US 2026/0245559 A1 · System And Method For Multi-Modal Podcast Summarization

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0245559 A1
- Document ID: US20260245559A1
- Status: Published application
- Kind: A1
- Application number: 19/178,773
- Filed: 2025-04-14
- Published: 2026-08-20
- Earliest claimed: 2020-07-15
- Assignees: Gracenote, Inc.
- Applicants: Gracenote, Inc.
- Inventors: Amanmeet Garg; Aneesh Vartakavi; Joshua Ernest Morris
- Web page: https://patentgrep.com/patent/US20260245559A1

## CPC

- G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking
- G10L 15/26: Speech to text systems (G10L15/08 takes precedence)
- G06F 16/64: Browsing; Visualisation therefor (generation of a list or set of audio data G06F16/638)
- G06F 40/216: using statistical methods
- G06F 40/295: Named entity recognition
- G06F 40/30: Semantic analysis
- G10L 15/02: Feature extraction for speech recognition; Selection of recognition unit
- G11B 27/031: Electronic editing of digitised analogue information signals, e.g. audio or video signals
- G10L 17/00: Speaker identification or verification techniques
- G10L 25/81: for discriminating voice from music
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G11B: INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER (recording measured values in a way that does not require playback through a transducer G01D9/00; recording or playback apparatus using mechanically marked tape, e.g. punched paper tape, or using unit records, e.g. punched or magnetically marked cards G06K; transferring data from one type of record carrier to another G06K1/18; circuits for coupling output of reproducer to radio receiver H04B1/20; gramophone pick-ups or like acoustic electromechanical transducers or circuits therefor H04R)

## Abstract

In one aspect, a method includes receiving podcast content, generating a transcript of at least a portion of the podcast content, and parsing the podcast content to (i) identify audio segments within the podcast content, (ii) determine classifications for the audio segments, (iii) identify audio segment offsets, and (iv) identify sentence offsets. The method also includes based on the audio segments, the classifications, the audio segment offsets, and the sentence offsets, dividing the generated transcript into text sentences and, from among the text sentences of the divided transcript, selecting a group of text sentences for use in generating an audio summary of the podcast content. The method also includes based on timestamps at which the group of text sentences begin in the podcast content, combining portions of audio in the podcast content that correspond to the group of text sentences to generate an audio file representing the audio summary.

## Claims

### Claim 1

A method comprising:
receiving podcast content;
generating a transcript of at least a portion of the podcast content;
parsing the podcast content to (i) identify audio segments within the podcast content, (ii) determine classifications for the audio segments, or (iii) detect pauses representative of transitions in audio of the podcast content;
based on the audio segments, the classifications, and the pauses, dividing the transcript into text sentences;
from among the text sentences of the transcript, selecting a group of text sentences for use in generating an audio summary of the podcast content; and
combining portions of audio in the podcast content that correspond to the group of text sentences to generate an audio file representing the audio summary of the podcast content.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
parsing the podcast content to identify speakers of the audio segments; and
labeling each text sentence of the transcript with a respective one of the speakers,
wherein selecting the group of text sentences for use in generating the audio summary of the podcast content is based at least in part on the speakers with which at least a portion of the group of text sentences are labeled.

### Claim 3 (depends on 1)

The method of claim 1, wherein the classifications distinguish speech audio segments from music audio segments.

### Claim 4 (depends on 1)

The method of claim 1, wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each being within a pre-determined time range within the podcast content.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
receiving a pre-determined text or audio provider-created summary associated with a content provider of the podcast content,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each having at least a threshold quantity of keywords from a stored list of keywords, the stored list of keywords including proper nouns and dates.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
receiving a pre-determined text or audio provider-created summary associated with a content provider of the podcast content,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each having at least a threshold degree of similarity with sentences from the pre-determined text or audio provider-created summary, and
wherein the group of text sentences each having at least a threshold degree of similarity with sentences from the pre-determined text or audio provider-created summary is based at least in part on the group of text sentences each having at least a threshold quantity of keywords that are included as text or spoken in the pre-determined text or audio provider-created summary.

### Claim 7 (depends on 1)

The method of claim 1, wherein the podcast content represents an episode of a podcast, the method further comprising:
receiving metadata associated with the podcast content, the metadata identifying a content provider associated with the podcast content, a title of the podcast, and a title of the episode,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the metadata.

### Claim 8 (depends on 1)

The method of claim 1, wherein the method is performed by a server computing system of a content distribution system that maintains and provides access to the podcast content.

### Claim 9 (depends on 1)

The method of claim 1, wherein parsing the podcast content comprises parsing the podcast content to identify the audio segments within the podcast content.

### Claim 10 (depends on 1)

The method of claim 1, wherein parsing the podcast content comprises parsing the podcast content to determine the classifications for the audio segments.

### Claim 11 (depends on 1)

The method of claim 1, wherein parsing the podcast content comprises parsing the podcast content to detect the pauses representative of the transitions in audio of the podcast content.

### Claim 12

A non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by a processor, cause performance of a set of operations comprising:
receiving podcast content;
generating a transcript of at least a portion of the podcast content;
parsing the podcast content to (i) identify audio segments within the podcast content, (ii) determine classifications for the audio segments, or (iii) detect pauses representative of transitions in audio of the podcast content;
based on the audio segments, the classifications, and the pauses, dividing the transcript into text sentences;
from among the text sentences of the transcript, selecting a group of text sentences for use in generating an audio summary of the podcast content; and
combining portions of audio in the podcast content that correspond to the group of text sentences to generate an audio file representing the audio summary of the podcast content.

### Claim 13 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, the set of operations further comprising:
parsing the podcast content to identify speakers of the audio segments; and
labeling each text sentence of the transcript with a respective one of the speakers,
wherein selecting the group of text sentences for use in generating the audio summary of the podcast content is based at least in part on the speakers with which at least a portion of the group of text sentences are labeled.

### Claim 14 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, wherein the classifications distinguish speech audio segments from music audio segments.

### Claim 15 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each being within a pre-determined time range within the podcast content.

### Claim 16 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, the set of operations further comprising:
receiving a pre-determined text or audio provider-created summary associated with a content provider of the podcast content,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each having at least a threshold quantity of keywords from a stored list of keywords, the stored list of keywords including proper nouns and dates.

### Claim 17 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, the set of operations further comprising:
receiving a pre-determined text or audio provider-created summary associated with a content provider of the podcast content,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the group of text sentences each having at least a threshold degree of similarity with sentences from the pre-determined text or audio provider-created summary, and
wherein the group of text sentences each having at least a threshold degree of similarity with sentences from the pre-determined text or audio provider-created summary is based at least in part on the group of text sentences each having at least a threshold quantity of keywords that are included as text or spoken in the pre-determined text or audio provider-created summary.

### Claim 18 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, wherein the podcast content represents an episode of a podcast, the set of operations further comprising:
receiving metadata associated with the podcast content, the metadata identifying a content provider associated with the podcast content, a title of the podcast, and a title of the episode,
wherein selecting the group of text sentences comprises selecting the group of text sentences based at least in part on the metadata.

### Claim 19 (depends on 12)

The non-transitory computer-readable storage medium of claim 12, wherein the processor is a processor of a server computing system of a content distribution system that maintains and provides access to the podcast content.

### Claim 20

A computing system comprising:
a processor; and
a non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by the processor, cause performance of a set of operations comprising:
receiving podcast content;
generating a transcript of at least a portion of the podcast content;
parsing the podcast content to (i) identify audio segments within the podcast content, (ii) determine classifications for the audio segments, or (iii) detect pauses representative of transitions in audio of the podcast content;
based on the audio segments, the classifications, and the pauses, dividing the transcript into text sentences;
from among the text sentences of the transcript, selecting a group of text sentences for use in generating an audio summary of the podcast content; and
combining portions of audio in the podcast content that correspond to the group of text sentences to generate an audio file representing the audio summary of the podcast content.
