人们没有足够的装备来检测人工智能驱动的语音克隆
Sarah Barrington1, Emily A Cooper2,3, Hany Farid4,5
1School of Information, University of California, Berkeley, CA, 94720, USA.
Scientific reports
|March 31, 2025
概括
人类听众很难区分人工智能 (AI) 产生的声音与真实的人类语言. 感知研究表明,人工智能声音具有高度现实性,因此很难准确识别它们.
科学领域:
- 语音处理 语音处理
- 人与计算机的互动.
- 人工智能的人工智能是人工智能.
背景情况:
- 生成型人工智能 (AI) 正在快速发展,通过各种方式产生越来越多的人类类内容.
- 人工智能产生的音频,特别是声音的真实性是持续研究和开发的关键领域.
研究的目的:
- 评估人工智能产生的声音的感知现实性.
- 评估人类参与者识别人工智能产生的声音并将其与原始发言人的声音相匹配的能力.
主要方法:
- 进行了一系列涉及人类参与者的感知研究.
- 介绍了参与者对人工智能产生的声音和人类产生的声音的录音.
- 收集了关于语音身份匹配和AI语音识别准确性的数据.
主要成果:
- 人类参与者表现出有限的能力,以始终识别AI产生的声音.
- 参与者认为人工智能产生的声音的身份与他们的真实对手在很大一部分时间是相同的.
- 准确的识别一个声音作为人工智能产生的发生只有大约[公式:见文本]的时间.
结论:
- 人工智能产生的声音表现出高度的现实主义,挑战人类的感知.
- 人工智能声音的不可区分性对数字通信的真实性和信任有重大影响.
- 需要进一步的研究来了解AI语音感知的细微差别,并开发强大的检测方法.
更多相关视频
09:09Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
Published on: September 27, 2024
381
06:24Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications
Published on: January 5, 2024
735
相关概念视频
Nonconscious Mimicry
4.5K
Nonconscious mimicry occurs when individuals alter their mannerisms to match the behaviors and expressions of those nearby, without intention.
4.5K
Masking and Demasking Agents
2.2K
EDTA titrations may necessitate masking and demasking agents to temporarily protect a particular metal ion in a mixture from the EDTA reaction. These agents facilitate the sequential analysis of the metal ions by forming stable complexes with some—but not all—metal ions during certain steps.
There are many masking agents, such as cyanide, fluoride, triethanolamine, thiourea, and 2,3-bis(sulfanyl)propan-1-ol (formerly 2,3-dimercapto-1-propanol), with the masking agent chosen based on...
There are many masking agents, such as cyanide, fluoride, triethanolamine, thiourea, and 2,3-bis(sulfanyl)propan-1-ol (formerly 2,3-dimercapto-1-propanol), with the masking agent chosen based on...
2.2K
Reproductive Cloning
29.8K
Reproductive cloning is the process of producing a genetically identical copy—a clone—of an entire organism. While clones can be produced by splitting an early embryo—similar to what happens naturally with identical twins—cloning of adult animals is usually done by a process called somatic cell nuclear transfer (SCNT).
Somatic Cell Nuclear Transfer
In SCNT, an egg cell is taken from an animal and its nucleus is removed, creating an enucleated egg. Then a somatic...
Somatic Cell Nuclear Transfer
In SCNT, an egg cell is taken from an animal and its nucleus is removed, creating an enucleated egg. Then a somatic...
29.8K
Hearing
51.4K
When we hear a sound, our nervous system is detecting sound waves—pressure waves of mechanical energy traveling through a medium. The frequency of the wave is perceived as pitch, while the amplitude is perceived as loudness.
51.4K
Cloning of Dolly the Sheep
3.2K
The first successfully cloned mammal was Dolly, a sheep, born on 5th July 1996 at Roslin Institute, Scotland. The cloned sheep was named after the American singer Dolly Parton. Dolly lived for seven years and died of respiratory complications, which is speculated to be due to the actual age of her DNA. Because the DNA in cloned cells belongs to an older individual, the cloned individual’s life expectancy may be affected. Indeed, analysis of Dolly’s DNA revealed shorter...
3.2K
Air-entraining Agents
68
Air-entraining agents improve the durability and workability of concrete in climates with frequent freezing and thawing. These agents prevent cracks by introducing small air bubbles into the mix, creating spaces accommodating water expansion when temperatures drop. The air-entraining agents lower the surface tension of water, forming stable, small air bubbles. This method is more effective than having accidental large voids, as the intentional, smaller, and evenly distributed air voids improve...
68
