How AI Predicts Who Will Speak Next in a Conversation
IBM's patent describes a system that uses neural networks to analyze speech patterns and intentions to predict which person will talk next in a conversation.
Original patent title: “End-of-turn detection in spoken dialogues”
IBM's patent describes a system that uses neural networks to analyze speech patterns and intentions to predict which person will talk next in a conversation. Granted to International Business Machines Corp in 2023 with 23 claims.
Key facts
Coverage
What does this patent actually cover?
This system uses a neural network to monitor a conversation between two people and figure out who is likely to speak next. It does this by simultaneously analyzing two things: the speaker's 'intention' (like whether they are asking a question or making a statement) and the 'turn type' (like whether they are trying to hold the floor or are about to switch speakers). By combining these predictions using a joint loss function, the system can determine if the first person will keep talking or if the second person will jump in. For example, if the system detects a rising pitch at the end of a sentence (an acoustic cue) and identifies the intention as a question, it can predict a turn switch to the other person.
The gap
What does this patent NOT cover?
- Does not cover systems that only look at text transcripts without analyzing acoustic cues like pitch or speaking rate.
- Does not cover simple rule-based systems that rely solely on silence duration to detect turn-taking.
- Does not cover systems that predict the content of the next sentence, only the identity of the next speaker.
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
What made this novel
The innovation lies in training a single neural network to jointly optimize two different tasks—intention recognition and turn-type detection—rather than treating them as separate, sequential steps.
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
Smart home voice assistants
Automated customer service phone bots
Real-time speech-to-text transcription software
Why it matters
The bigger picture
Managing natural conversation flow is a major hurdle for voice assistants like Siri or Alexa, which often interrupt users or stop listening too early. By accurately predicting turn-taking, this technology aims to make human-computer interactions feel more fluid and less robotic. It is part of a broader effort to move beyond simple keyword recognition into true conversational understanding.
Filed
December 23, 2020
Granted
May 9, 2023
Market context
Who's building on this
Companies in this space
IBM remains a primary player in enterprise AI and natural language processing. Other major tech companies like Google, Amazon, and Microsoft are actively researching similar conversational AI models to improve the responsiveness of their voice-enabled products.
Market impact
This technology contributes to the ongoing refinement of conversational AI, aiming to reduce the latency and awkwardness inherent in current voice interfaces. It supports the industry trend toward more human-like, context-aware digital assistants in both consumer and enterprise sectors.
Claim 1 — Plain English
What this patent covers
This system uses a neural network to monitor a conversation between two people and figure out who is likely to speak next. It does this by simultaneously analyzing two things: the speaker's 'intention' (like whether they are asking a question or making a statement) and the 'turn type' (like whether they are trying to hold the floor or are about to switch speakers). By combining these predictions using a joint loss function, the system can determine if the first person will keep talking or if the second person will jump in. For example, if the system detects a rising pitch at the end of a sentence (an acoustic cue) and identifies the intention as a question, it can predict a turn switch to the other person.
The clever bit
The innovation lies in training a single neural network to jointly optimize two different tasks—intention recognition and turn-type detection—rather than treating them as separate, sequential steps.
What it does not cover
- Does not cover systems that only look at text transcripts without analyzing acoustic cues like pitch or speaking rate.
- Does not cover simple rule-based systems that rely solely on silence duration to detect turn-taking.
- Does not cover systems that predict the content of the next sentence, only the identity of the next speaker.
Patent timeline
Application submitted to the patent office
Application published, typically 18 months after filing
Patent officially issued
PatentBrief Score
Impact Score
Early stage
Citation count
0/40
No citations yet
Claim breadth
15/20
Broad claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →
Recency
20/20
Granted within 5 years
Assignee scale
0/20
Independent or smaller assigneeassigneeThe entity that owns the patent — usually the inventor's employer or a company.Read more →
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$47K – $150K
Midpoint $94K · 14.5 yr remaining · industry ×1.6
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
The original legal language
Original claims
23 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Provost, E. M., Polymenakos, L., Aldeneh, Z., & Dimitriadis, D. B. (2023). How AI Predicts Who Will Speak Next in a Conversation (U.S. Patent No. 11,645,473). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/11645473/palm-pathways-language-model
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US11645473"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
More to explore
More in AI & Machine Learning
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11615208 · 2023 · Capital One Services LLC
How Cloud Systems Automatically Create and Train AI Data Models
US 10402750 · 2019 · Facebook Inc
How Facebook Uses Deep Learning to Predict What You Might Like
New to patents?
Common Questions
Frequently Asked Questions
What does How AI Predicts Who Will Speak Next in a Conversation cover?
IBM's patent describes a system that uses neural networks to analyze speech patterns and intentions to predict which person will talk next in a conversation.
Who owns patent US 11645473?
International Business Machines Corp owns this patent, granted in 2023.
When does this patent expire?
This patent is expected to expire on May 9, 2043, when the invention enters the public domain.
What problem does this patent solve?
Managing natural conversation flow is a major hurdle for voice assistants like Siri or Alexa, which often interrupt users or stop listening too early. By accurately predicting turn-taking, this technology aims to make human-computer interactions feel more fluid and less robotic. It is part of a broader effort to move beyond simple keyword recognition into true conversational understanding.
What does this patent NOT cover?
Does not cover systems that only look at text transcripts without analyzing acoustic cues like pitch or speaking rate.
Same assignee
More from International Business Machines Corp
Patent monitoring