While CES seems to be very quiet this year, announcements were schedule anyway. As expected there are a few that feature Artificial Intelligence, or more likely Machine Learning. Hat tip to Phil Lelyveld of the Entertained Technology Center for the reporting.
Audio Design Desk is an AI-assisted content creation tool that reinvents the process of connecting audio to video. It is particularly good with sound design, foley-like sound effects and music. It will make recommendations based on the type of music and effects that have been imported, and intelligent replace sounds without losing sync. I’d also note that it’s another “helper” tool designed to amplify the creativity of those working with it.
Supertone is an AI-based audio tech startup that is interesting because it can capture an actor or singer’s tone and rich expression and separate out voices from background. Separating voices is not really new, but capturing the tone an expression from a performance, and applying it to another voice, perhaps even an artificial one. The example cited was to apply the original performance attributes to new (dubbed) dialog for international distribution.
It’s a milestone to having a library of performances that can be applied to any artificial character or voice.
While they are the AI highlights, there are a lot of very interesting new technologies highlighted in the article above, and the followup for day two.