Apps
Games
Phones
Recently,
OpenAI has made strides in the development of voice imitation technology. In
their latest update, they explained how their Voice Engine, first revealed
towards the end of 2022, has evolved. Presently, it's possible for this system
to replicate any voice with just a brief 15-second audio clip.
The
organization has created a voice AI system that's really impressive. It can
make sound versions that sound just like a person speaking, and it only needs a
little bit of the person's voice to do it. This cool method could help in many
ways, like helping people read, translating between different languages, and
giving a voice to those who have trouble speaking.
The firm's
innovation works quite straightforwardly: individuals just have to supply an
audio clip lasting 15 seconds along with a text script. Following this, the
artificial intelligence can mimic the person's voice to convey the script's
message in a way that sounds genuine and can capture emotions.
Led by Sam
Altman, the organization is aware of the dangers that might come with easy
access to such technology and promotes careful consideration and knowledge.
They stress the necessity of carefully evaluating the potential hazards that
might arise from misusing this tech, drawing lessons from past frauds where
voice imitation was used to trick people.
Right now,
this new tech isn't available for everyone, and the company hasn't shared if
they plan to make it as publicly available as ChatGPT. They believe it's key to
run a few tests first to see if it's practical to add this technology to their
products.
