Scientists Develop Software to Alter Recordings of Human Voice

, , Comments Off on Scientists Develop Software to Alter Recordings of Human Voice

Researchers have designed a new software that permits people to revise human voice’s audio recording with the simplicity of altering words on a screen of computer. The technology is developed by scientists in the U.S. at Princeton University and can do those things for human voice’s audio recordings what software of word processing did for the written word.

Scientists Develop Software to Alter Recordings of Human Voice

The software, dubbed VoCo, offers a simple means to replace or add a word in an human voice’s audio recording by altering a transcript of the soundtrack. New words are synthesized automatically in the voice of speaker even though they do not emerge anywhere else in the soundtrack. The system, which employs a complicated algorithm to recreate and learn the sound of a specific voice, could one day make altering narration and podcasts in videos much simpler.

The technology can offer a turning point for making personalized voices of robots that seem to be natural. “VoCo offers a glance at a very specific technology for altering audio recordings, but it is also an indication for technologies in future that will permit the human voice to be automated and synthesized in extraordinary ways,” said Professor of Computer Science at Princeton, Adam Finkelstein.

User interface of VoCo seems similar to other audio altering software on a computer screen such as the Audacity, the famous podcast editing program, or GarageBand, music editing program of Apple. It provides visualization of the waveform of the audio recording and a group of copy, cut, and paste tools for altering. Dissimilar to other programs, VoCo also enhances the waveform with the help of a text transcription of the recording and permits the user to insert new words or replace words, which are not already present in the recording, just by typing in the transcription.

As soon as the user types the new word, VoCo upgrades the audio recording, robotically synthesizing the new word by edging snippets of audio together from somewhere else in the narration.

In short, technology is making true those dreams that were just an imagination in movies some years ago. And we hope that the coming years will bring some more such inventions.