A report by Timothy B Lee shared how he recently put his A.I. cloning skills to the test, asking family and strangers to guess which of his audio clips were computer-generated and which were real. While the results weren't great for his self-esteem, they indicated that A.I. is improving at replicating the human voice.
A Blind Test for Lee's Cloned Voice: Can His Family Tell the Difference?
According to the story by Slate, using software from IBM and Python programming, he created 12 audio clips with the first 12 paragraphs of an article that Lee wrote. Seven clips were his natural voice, and five were generated by A.I., with no one knowing which was which.
The researcher asked family and strangers to identify which clips were real and which weren't. His family members had a better track record than the broader internet, with Lee's mother, wife, and siblings correctly calling seven out of the 12 paragraphs of his cloned voice. The article reportedly came from Understand AI, which looks at the impact of artificial technology in the world.
Voice Cloning: Results from Experiments Show an Increase in Results and Quality
A grad school friend and a former employee weren't as successful, correctly guessing only 11 and 10 out of 12 times, respectively. Results from the broader internet were even worse, with anonymous listeners only correctly guessing 54 percent of the time.
Voice cloning has come a long way in just a few years. Cloning a voice is much easier and cheaper than a few years ago. At the same time, the quality has improved to be convincing enough that the average person can't reliably tell the difference between cloned and real voices.
The Possibilities of Voice Cloning: Revolutionizing Communication and Opening Creative Possibilities
This has the potential to revolutionize many industries and to create a whole new way for people to communicate-to have their voices speak for them long after they are gone, for their ideas to be heard, and to be passed on to the next generation.
The technology behind voice cloning is transforming how we communicate and even finding uses in more creative areas. The fake video of Richard Nixon, created by MIT researchers and Respeecher, is an excellent example of what can be done with this technology.
Read Also: Amazon Releases Generative AI Suite, Titan and Bedrock, Soon Coming to AWS
The Privacy and Security Concerns of Voice Cloning Technology
Companies like Respeecher are making technologies like A.I. and speech synthesis accessible to filmmakers and artists, opening up a world of possibilities for how stories can be told.
At the same time, the technology behind voice cloning carries privacy concerns. Everyone now lives in a world where it is possible to imitate anyone's voice, so it's important to be aware of potential privacy and security concerns.
As this technology continues to improve, regulators, software developers, and consumers must be aware of the potential risks and work together to ensure these technologies are used responsibly.
Related Article: ChatGPT Used to Create Dangerous Data-Stealing Malware, Researcher Claims
