Skip to main content

ByteDance Unveils StreamVoice: AI-Powered Live Voice Conversion Raises Deepfake Concerns and Misinformation Risks

 ByteDance, the renowned Chinese technology firm responsible for the popular TikTok platform, has unveiled something new for its users—StreamVoice. This tool, leveraging generative-AI technology, enables users to seamlessly alter their voices to mimic others.


As of now, StreamVoice remains inaccessible to the general public, yet its introduction underscores the noteworthy progress in AI development. The tool facilitates the effortless creation of audio and visual impersonations of public figures, commonly referred to as "deepfakes." Notable instances include the use of AI to emulate the voices of President Joe Biden and Taylor Swift, a phenomenon particularly prevalent as the 2024 election looms.

Collaborating on this groundbreaking initiative are technical researchers from ByteDance and Northwestern Polytechnical University in China. It's imperative to note that Northwestern Polytechnical University, recognized for its collaborations with the Chinese military, should not be confused with Northwestern University in the United States.

In a recently published paper, the researchers underscore StreamVoice's capacity for "real-time conversion" of a user's voice to any desired alternative, requiring only a singular instance of speech from the target voice. The output unfolds at livestreaming speed, boasting a mere 124 milliseconds of latency—a significant achievement in light of historical limitations associated with AI voice conversion technologies, traditionally effective in offline scenarios.

The researchers attribute StreamVoice's success to recent advancements in language models, enabling the creation of a tool that performs live voice conversion with high speaker similarity for both familiar and unfamiliar voices. Experiments, as detailed in the paper, emphasize the tool's efficacy in streaming speech conversion while maintaining performance comparable to non-streaming voice conversion systems.

Referring to Meta's Llama large language model, a prominent entity in the AI landscape, the paper details the utilization of the "LLaMA architecture" in constructing StreamVoice. Additionally, the researchers incorporated open-source code from Meta's AudioDec, described by Meta as a versatile "plug-and-play benchmark for audio codec applications." Training primarily on Mandarin speech datasets and a multilingual set featuring English, Finnish, and German, the researchers achieved the tool's proficiency.

Although the researchers refrain from prescribing specific use cases for StreamVoice, they acknowledge potential risks, such as the dissemination of misinformation or phone fraud. Users are encouraged to report instances of illegal voice conversion to appropriate authorities.

AI experts, cognizant of advancing technology, have long cautioned against the escalating prevalence of deepfakes. A recent incident involved a robocall deploying a deepfake of President Biden, urging people not to vote in the New Hampshire primary. Authorities are currently investigating this deceptive robocall, underscoring the urgent need for vigilance in the face of evolving AI capabilities.

Content generated using AI and reviewed by humans. Photo: DIW - AIGen
Labels:

Comments

Post a Comment

Popular posts from this blog

Is Google Sunsetting Gmail In 2024? New Hoax Leaves Users Confused

  Rumors are going strong on social media about Google’s journey with Gmail coming to an end. And that too as early as this year. If you paid witness to such a shocking hoax that’s fast spreading online too then we can confirm that it’s not true by any means. The Android maker is addressing the need for users to calm down after the hoax emphasizing the company’s goodbye to Gmail. Many versions of this hoax can be found online  on the Twitter (X) app . But Google has rest assured everyone to take it easy and that it’s not going anywhere. However, a clause also mentions how users would now be defaulted to new views on Gmail. So other than those cases where you had specifically requested for the basic HTML version, nothing is going to change. The search engine giant  just made  updates to its respective  support page  too instead of this hoax and how it will now change to Standard display after the issuance of a deadline date. Reports by users on tech outlet...
Post a Comment
Read more

Telegram is Home to Many Scammers and Cyber Criminals Who Sell Many Phishing Tools at Cheap Prices

  Telegram is a popular messaging app that has over one billion downloads and 700 million monthly users. However, this popular app is quite known for its selling and buying of phishing kits that scam users and sell their harmful wares. A cybersecurity startup, Guardio, published  a report  that sheds light on some phishing tools on Telegram. It was revealed that there are some harmful kits available on Telegram that can easily hack users’ social media and bank accounts logins. Guardia said that Telegram is a scammer’s heaven due to easy access and cheap prices of many phishing tools. There are also many phishing groups on Telegram where different scammers share data with other scammers and use phishing tools for hacking and phishing. For a few years, Telegram has become a home to many seasonal scammers and newcomers who sell malicious tools to steal the data of people. Some of the expensive phishing tools on Telegram also let the scammers create fake pages that easily byp...
Post a Comment
Read more

200,000 Private Records From Facebook’s User Database Stolen, Hackers Forum Confirms

  A hacker has just raised the alarm amongst Facebook users after confirming that 200,000 personal data record entries were reportedly stolen from the firm’s database. The news is alarming for obvious reasons as the claims further went on to delineate how the cybercriminal dubbed ‘alogoatson breached contractors’ efforts that are in charge of Facebook’s cloud services. They stole part of the user database that featured a significant number of entries. The information was rolled out by a leading threat actor dubbed ‘IntelBroker’ which is notorious for a long list of leaks that entailed data stolen through General Electric and a long list of high-profile attacks taking place. This sample entails lists featuring full names, profile image links, and hashed passwords. Other than that, profile ratings, settings, and plenty of reviews were on display. The hacker explained how the data that was compromised included the likes of Physical IDs. This database was first rolled out in February a...
Post a Comment
Read more