NASA’s new dark energy space telescope can also detect killer asteroids
technologyreview.com·3h ago
AI Research
arXiv CS.AI·May 28·5 min read
I Hear, Therefore I Trust: A Socio-Technical Investigation of Humans as Synthetic Speech Detectors
TL;DR
As synthetic speech becomes more prevalent, there is a growing concern about humans' ability to detect it. This study developed a framework to evaluate human performance in identifying synthetic speech using a series of listening tests.
✦ Why It Matters
Engineers can leverage these findings to create better synthetic speech detection tools and improve user training.
Key Takeaways
- Explore the impact of trust cues in synthetic speech applications.
- Train users to recognize different types of synthetic speech.
- Utilize quality ratings to improve speech synthesis systems.
- Investigate further into socio-technical factors affecting detection.
How It Works
Participants marked suspected synthetic segments in a task that tested their ability to discern between authentic and synthetic speech. The study manipulated trust cues to see how they influenced detection behavior, revealing that while these cues did not improve accuracy, they affected participants' engagement in the detection process.