Seoul Gazette - OpenAI unveils voice-cloning tool

NYSE - LSE

RBGPF	4.34%	65.86	$
RYCEF	4.06%	10.6	$
CMSC	-0.23%	22.11	$
VOD	-1.62%	9.25	$
RELX	-1.5%	54.06	$
GSK	-0.81%	36.87	$
RIO	-1.42%	59.18	$
NGG	-3.41%	70.18	$
BTI	-2.66%	43.3	$
BP	1.61%	28.59	$
SCS	5.44%	10.48	$
BCC	2.77%	89.58	$
CMSD	-0.36%	22.33	$
JRI	-0.59%	12.95	$
BCE	4.41%	22.23	$
AZN	-4.12%	67.3	$

OpenAI unveils voice-cloning tool

TECHNOLOGY 29.03.2024

OpenAI unveils voice-cloning tool / Photo: © AFP

OpenAI on Friday revealed a voice-cloning tool it plans to keep tightly controlled until safeguards are in place to thwart audio fakes meant to dupe listeners.

Text size:

A model called "Voice Engine" can essentially duplicate someone's speech based on a 15-second audio sample, according to an OpenAI blog post sharing results of a small-scale test of the tool.

"We recognize that generating speech that resembles people's voices has serious risks, which are especially top of mind in an election year," the San Francisco-based company said.

"We are engaging with U.S. and international partners from across government, media, entertainment, education, civil society and beyond to ensure we are incorporating their feedback as we build."

Disinformation researchers fear rampant misuse of AI-powered applications in a pivotal election year thanks to proliferating voice cloning tools, which are cheap, easy to use and hard to trace.

Acknowledging these problems, OpenAI said it was "taking a cautious and informed approach to a broader release due to the potential for synthetic voice misuse."

The cautious unveiling came a few months after a political consultant working for the long-shot presidential campaign of a Democratic rival to Joe Biden admitted being behind a robocall impersonating the US leader.

The AI-generated call, the brainchild of an operative for Minnesota congressman Dean Phillips, featured what sounded like Biden's voice urging people not to cast ballots in January's New Hampshire primary.

The incident caused alarm among experts who fear a deluge of AI-powered deepfake disinformation in the 2024 White House race as well as in other key elections around the globe this year.

OpenAI said that partners testing Voice Engine agreed to rules including requiring explicit and informed consent of any person whose voice is duplicated using the tool.

It must also be made clear to audiences when voices they are hearing are AI generated, the company added.

"We have implemented a set of safety measures, including watermarking to trace the origin of any audio generated by Voice Engine, as well as proactive monitoring of how it's being used," OpenAI said.

B.Ha--SG

Featured

Germany slams Russian 'lies', France warns of war 'spectre' in WWII commemorations

Germany on Thursday slammed Russia's "lies" over its invasion of Ukraine, while French leader Emmanuel Macron warned that the "spectre of war" has returned to Europe as western allies staged solemn ceremonies to mark 80 years since the victory over the Nazis.

UK Certifies Protector As First of Its Kind Remotely Piloted Aircraft

Obtaining the MTC has been a goal of GA-ASI since the inception of the MQ-9B in January 2014. The company took its proven UAS platform, the MQ-9A, and added performance enhancing features while ensuring that the design was capable of meeting NATO's STANAG 4671 Edition 2 airworthiness requirements. To meet those rigorous requirements, the aircraft incorporates numerous enhancements not found on other large UAS. These include lightning protection, fire protection, anti-icing systems and a fatigue-and-damage-tolerant building block design approach. All flight critical software was designed in compliance with the rigor of Do-178/254. Mission software is rigidly separated from flight critical software. These features not only address the aircraft's airworthiness but also enhance its reliability and operational flexibility to levels unmatched by other UAS.

Malaysia Cybersecurity Center of Excellence Marks First Anniversary with New Partnerships, Scholarships and Expanded Programs

BlackBerry expands CCOE Curriculum with free cybersecurity training for women, students, government and low-access communities, fostering a more inclusive and skilled cyber workforce

Meta blocks access to Muslim news page in India

Meta has banned a prominent Muslim news page on Instagram in India at the government's request, the account's founder said Wednesday, denouncing the move as "censorship" as hostilities escalate between India and Pakistan.