The past year has seen the emergence of artificial intelligence (AI) tools that can create images, artwork, or even video using a text prompt.
There have also been great strides forward AI typingwith ChatGPT from OpenAI It caused widespread excitement – and fear – for the future of writing.
Now, just a few days into 2023, another powerful AI use case is coming into the spotlight – a text-to-voice tool that can accurately simulate a person’s voice.
Developed by Microsoft, VALL-E can record a person’s voice for three seconds, loop that voice, and convert written words into speech, with realistic tone and emotion depending on the context of the text.
Trained with 60,000 hours of speech recordings in English, he can deliver a speech in a “pointless situation,” meaning no previous examples or training in a given context or situation.
Introducing VALL-E in Paper published by Cornell UniversityThe developers explained that the recording data consists of more than 7,000 unique speakers.
The team says the TTS used hundreds of times more data than existing TTS systems, which helped them overcome the problem of non-payment.
The tool isn’t currently available for public use — but it does raise questions about safety, given that it can be used practically to generate any text emanating from anyone’s voice.
Microsoft is betting big on artificial intelligence
However, he lost I made an offerdisplays a number of three-second speaker prompts and an explanation of the text-to-speech process in action, properly simulating the voice.
Along with the speaker prompt and VALL-E output, you can compare the results with the “ground truth” – the actual speaker reading instant text – and the “ground” result from existing text-to-speech technology.
Microsoft has invested heavily in artificial intelligence and is one of the backers of OpenAI, the company behind ChatGPT and DALL-E, a text-to-image or art tool.
The software giant invested $1 billion (930 million euros) in OpenAI in 2019, and a report this week on semafor.com said it was looking to invest another $10 billion (9.3 billion euros) in the company.
“Web specialist. Lifelong zombie maven. Coffee ninja. Hipster-friendly analyst.”
The SteamWorld Series Creator admits the closing of the 3DS eShop is a little “annoying”.
The note says Google’s reorganization in Assistant after the launch of Bard
Zonai Legend of Zelda theories are back after the new TOTK trailer