Recently, OpenAI took its artificial intelligence (AI) powered ChatGPT bot to public testing and it took the internet by storm.
People are impressed with ChatGPT’s smart response to queries and many describe it as a huge leap forward in technology. However, he scared Google into improving his search engine before ChatGPT could hurt his business.
Impressed by the ingenuity of ChatGPT, Microsoft plans to invest $10 billion in OpenAI led by Sam Altman. Reportedly, the company is implementing ChatGPT technology in its latest version of its Edge browser and Bing search apps.
Now the Redmond-based company has announced VALL-E, an AI-based neural coding language model that can reproduce anyone’s voice and, most interestingly, only requires three seconds of audio recording. .
Just give a text, and it will easily speak with a specific person’s voice. It is said that a VALL-E speech can match the timbre and even the emotional tone of the speaker, as well as the acoustic environment of a room.
must read | ChatGPT: “code red” for Google’s search activity
“During the pre-training phase, we scaled the TTS (Text-to-Speech Synthesis) training data to 60,000 hours of English speech, which is hundreds of times greater than current systems. VALL-E highlights in-context learning capabilities and can be used to synthesize high-quality personalized speech With just 3 seconds of recording an invisible speaker as a voice prompt.
Currently, Microsoft will not release the VALL-E source code yet, as it could be misused.
Even the ChatGPT research paper revealed that people can use it to spread propaganda and misinformation.
Microsoft will test VALL-E more rigorously to ensure there is less room for abuse.
So don’t hold your breath. It will take some time before VALL-E falls into the public domain.
Get the latest news on new launches, gadget and app reviews, cybersecurity and more on personal technology only on Tech DH.