Grady College faculty member explores the future of dubbing

A new study, conducted by Laurena Bernabo, Grady College assistant professor in the Department of Entertainment and Media Studies, highlights dubbing professionals’ perspectives on the changing role of artificial intelligence (AI) in the dubbing industry. 

In Bernabo’s publication, “How, When, and Why to Use AI: Strategic Uses of Professional Perceptions and Industry Lore in the Dubbing Industry,” she explores developments in dubbing from human-centered practices to where neural machine translations and artificial intelligence are used.

Dubbing is the act of replacing the original dialogue in a film, television show or video with a recorded voice track to change the language of the piece. Producers must first write a script for the original piece in the language they wish to change it to. Then, voice actors and sound engineers work to match lip movements, emotions and tones of the original characters.

However, Bernabo shares that because of a lack of representation, many dubbing industries are not able to match the exact racial, ethical or sexual background of a character. One example is the dubbing of Pixar’s Soul in Denmark, in which the Black protagonist, originally voiced by Jamie Foxx, was dubbed by a white dubbing actor and thus failed to fully capture the character’s race.

Because of these discrepancies, some dubbing industries have looked to AI to improve accuracy, inclusivity and synchronization by generating voices that better match the original actors’ tone, emotion and cultural identity. Additionally, AI provides the dubbing industry a more affordable way to produce lower quality versions of lesser-known pieces.

However, Bernabo states it is a fine line trying to determine how and when can AI be utilized for dubbing. While AI is great for trying to match cultural identity, it struggles to create emotion and tone in productions. Because of this, many industry professionals prefer to use AI to write the dub script, while still utilizing people to do the voice recordings. Additionally, many viewers aren’t interested in AI voices in their content. For example, Spanish viewers are used to watching Tom Cruise movies and hearing him dubbed by voice actor Arturo Mercado. While AI could essentially replicate Tom Cruise’s voice and change the dialogue to Spanish, viewers in Spain express disdain for this possibility, insisting that they like ‘their’ Tom Cruise just fine.

When asked about the future of the dubbing industry, Bernabo shared, “The dubbing industry, like a living organism, is constantly evolving. Some of the affordances provided by AI feel revolutionary in their ability to democratize media and better communicate people’s identities in a global media landscape, but others are cost-cutting measures that sacrifice quality for the bottom line. I hope dubbing professionals continue to advocate for the cultural value of human labor and that this value is recognized as necessary for meaningful translations.”

Author: Lauren A. Pike, lauren.pike@uga.edu