Talker change detection: A comparison of human and machine performance
Neeraj Kumar Sharma1, Shobhana Ganesh2, Sriram Ganapathy2
1Department of Psychology, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 15213, USA.
The Journal of the Acoustical Society of America
|February 4, 2019
Summary
Human talker change detection (TCD) in conversations is challenging. This study shows humans outperform machines in identifying voice changes, with reaction times predictable by acoustic feature models.
Area of Science:
- Speech processing
- Psychoacoustics
- Human-computer interaction
Background:
- Automatic analysis of conversational audio is complex due to multiple speakers, intonation variations, and overlapping speech.
- Prior research predominantly focused on single-talker speech or the cocktail party effect, with limited investigation into talker change detection (TCD).
- Acoustic features significant for characterizing a talker's voice in conversational settings are underexplored.
Purpose of the Study:
- To examine human talker change detection (TCD) in multi-party speech.
- To investigate the acoustic features influencing TCD.
- To compare human TCD performance against state-of-the-art machine TCD systems.
Main Methods:
- A behavioral paradigm where listeners indicated perceived talker changes in multi-party speech utterances.
- Modeling acoustic feature distance between speech segments before and after a talker change.
- Evaluating model performance based on speech segment duration prior to change.
- Benchmarking human performance against online and offline machine TCD systems.
Main Results:
- Human reaction times in TCD tasks can be accurately estimated by acoustic feature distance models.
- Model estimation accuracy improves with longer preceding speech durations.
- Human TCD performance surpasses that of current state-of-the-art machine TCD systems.
Conclusions:
- Human listeners are adept at detecting talker changes in multi-party conversations.
- Acoustic feature analysis provides a viable method for modeling human TCD.
- Further research into TCD could enhance automatic speech processing systems.
Related Concept Videos
02:57The Sense of Self: Reflected Self-Appraisal and Social Comparison
56.0K
According to Charles Cooley, we base our image on what we think other people see (Cooley 1902). We imagine how we must appear to others, then react to this speculation. We don certain clothes, prepare our hair in a particular manner, wear makeup, use cologne, and the like—all with the notion that our presentation of ourselves is going to affect how others perceive us. We expect a certain reaction, and, if lucky, we get the one we desire and feel good about it. But more than that, Cooley...
56.0K
01:50Global Climate Change
28.9K
Throughout its ~4.5 billion year history, the Earth has experienced periods of warming and cooling. However, the current drastic increase in global temperatures is well outside of the Earth’s cyclic norms, and evidence for human-caused global climate change is compelling. Paleoclimatology, the study of ancient climate conditions, provides ample evidence for human-caused global climate change by comparing recent conditions with those in the past.
28.9K
01:19Machines
577
Machines are complex structures consisting of movable, pin-connected multi-force members that work together to transmit forces. One example of a machine is the cutting plier, which is used to cut wires by applying forces to its handles. When equal and opposite forces are exerted on the handles of the cutting plier, they cause the cutting edges to come together and apply equal and opposite reaction forces on the wire, which are greater than the applied forces.
A free-body diagram of the...
A free-body diagram of the...
577
01:30Machines: Problem Solving II
666
Machines are complex structures consisting of movable, pin-connected multi-force members that work together to transmit forces. Consider a lifting tong carrying a 100 kg load. It comprises movable sections DAF and CBG linked together with member AB.
666
01:22Net Change Theorem
69
The Net Change Theorem is a fundamental principle in calculus that establishes a direct relationship between a function’s rate of change and its accumulated change over an interval. Mathematically, it states that the definite integral of a function's derivative over a given interval [a,b] yields the net change in the original function:This theorem has significant applications in various real-world scenarios, including physics, economics, and engineering. A particularly useful application...
69
01:20Rates of Change
87
The rate of change is a central concept in mathematics that quantifies how one variable varies in response to another. It serves as a foundational tool in modeling dynamic systems across disciplines such as physics, biology, economics, and engineering. Understanding both average and instantaneous rates of change enables the analysis of behavior in functions that describe real-world phenomena.Average Rate of ChangeFor a function f(x) defined over an interval [x1,x2], the average rate of change...
87