Human or AI? I cloned my voice and asked friends if they could tell the difference
Assoc Prof Terence Sim from NUS’s School of Computing highlights the challenges posed by AI-generated voice fraud. He notes that the Singaporean accent can be harder to clone, as most AI models are trained on American or British accents. While skills such as spotting hissing, mismatched background sounds, or abrupt transitions in speech can help detect deepfake voices, he cautions that these methods are not foolproof as voice cloning technology continues to advance.
Trending Posts
13 February 2026
Lianhe Zaobao reported on the Government’s AI initiatives announced in Budget 2026, including the establishment of a National AI Council ...
29 April 2018
14 February 2019
6 March 2026
Tech in Asia reported on NUS School of Computing's collaboration with OpenAI to integrate Codex and other tools from OpenAI's ...
30 December 2018
20 May 2024
The National University of Singapore School of Computing (NUS Computing) has unveiled a new Computing Gallery, the first of its ...
22 May 2026
Associate Professor Harold Soh from the Department of Computer Science was quoted in South China Morning Post on Singapore's push ...
20 January 2021
26 February 2018
15 March 2019
26 June 2024
Professor Jungpil Hahn works for the Department of Information Systems and Analytics at the School of Computing, National University of ...
23 November 2021
12 September 2018
17 March 2019
14 January 2020
26 February 2024
Prof Anthony Tung highlights the critical link between legislation and technology, particularly in regulating AI. ...