Automated and synthetic voices found in over a quarter of unwanted calls
The Machines Are Calling: Measuring Automated and Synthetic Voices in Unwanted Inbound Calls
Cryptography and SecurityComputers and SocietySound
Summary
Many unwanted calls in the U.S. come from machines, but it has been unclear how often they use synthetic (computer-generated) voices versus recorded human voices. The authors set up a system that answered thousands of calls with AI personas and analyzed how the callers spoke. They found that at least about 27% of calls used machine voices—either synthetic speech or repeated recordings—and another 10% were silent, likely automated. Synthetic voices appeared mostly in spam calls trying to sell leads, and some callers revealed they were automated.
What this means in practice
- •For telecom fraud investigators: Identify and quantify the use of synthetic and replayed machine voices in unwanted call campaigns to better target enforcement and mitigation efforts.
- •For call center security teams: Detect incoming calls likely to use automated or synthetic voices to improve screening for spam and fraud before engaging agents.
Authors
Xingyu Shen, Tommy Duong, Muduo Xu, Xiaodong An, Jiaqi Gan, Haoyuan Tang, Jamey Z. Liang, Siyu Zhang, Yan Zhang, Simiao Ren
Abstract
In February 2024 the U.S. Federal Communications Commission (FCC) placed AI-generated voices under the Telephone Consumer Protection Act (TCPA). Yet no peer-reviewed measurement says how much unwanted call traffic is placed by a machine, or how much of that machine speech is synthesized rather than played from a recording. We report both with a disclosed pipeline. An interactive voice honeypot (language-model personas on real U.S. numbers, the caller recorded on its own track) recorded 10,987 calls over 66 days; 11 days on which our stack answered silently are set aside. Three instruments read each opening: an audio fingerprint that finds the same recording played on other calls, a commercial synthetic-speech detector on the caller's first ten seconds, and blinded listeners who check what it flags. Of the 7,233 calls our persona greeted on normal days, 13.8% open with a recording we also heard on another call, and 13.1% with fresh audio the detector labels synthetic. A further 9.9% open with a caller who never spoke after our greeting, 54.2% with fresh audio the detector labels human, and 9.0% could not be scored. Machine-voiced openings are therefore at least 26.9%, a further tenth of calls are silent connections we read as machine-placed, and replays of a recording make up 45% of the detector's own rate (29.3% of 6,192 scored openings). The same waveform played on two calls lands on opposite sides of the detector's threshold 13.6% of the time, and eleven listeners confirm 54.4% of what it flags. Synthetic openings concentrate in lead-generation spam (33.8%), not fraud (21.1%); 0.44% disclose automation. Prevalence tracks how long a bait number has circulated (59% against 19% in the same weeks): seeding history, not calendar time, explains the trend. Campaigns outlast their numbers: one recorded compliance notice opens calls in six campaigns, and one synthetic voice serves nine.