Prime Video's AI dub is the worst. So voice actors are furious.

The artificial intelligence efforts are not working out well for Prime Video’s Korean dramas.

We have been hearing for a few weeks about how the use of artificial intelligence is likely to become an essential component in the process of enhancing and elevating some areas of audiovisual creation to levels that have never been seen before. A recent example of the outstanding use of artificial intelligence to develop effects that make the actors Anya Taylor-Joy and Alyla Browne, who play the protagonist of ‘Furiosa’ at different ages, appear alike is something that we have seen. However, the luck of each and every example is not the same.

AI in dubbing. Despite the fact that dubbing is a significant part of the film and series industry in countries where English is not the primary language, new circumstances have arisen as a result of the globalization of content brought about by the proliferation of streaming services. The public no longer only need information to be presented in English; languages such as Korean and Spanish are now often used on platforms that originate from a variety of countries. ‘The Squid Game‘ is the undisputed example of this, as it continues to be the most watched series in the history of Netflix, and it is produced in Korean.

Why should you go with AI? The problem is that not all languages are as in demand as Korean or Spanish, and there are minority languages that would require an exorbitant financial investment due to the fact that translation, dubbing, quality control, and other requirements are all necessary. Because artificial intelligence makes it feasible to save in this manner and reach places that would otherwise remain undubbed, businesses like Deepdub, Papercup, Resemble AI, and ElevenLabs are developing to provide dubbing services using AI. These companies sell their services to customers.

The abrupt change in lip movement. Using generative artificial intelligence to adjust the lips of the interpreters to different languages or, for example, to script adjustments to adapt the video to different ages of the audience is the next step that will be taken. It is a subject that already has connections to other issues, such as the use of deepfakes in order to revitalize performers. According to the findings of a survey conducted by Variety, agencies that specialize in the subject matter, such as LipDub, are expanding their clientele on platforms that are less demanding, such as YouTube, where they already have one hundred clients, but are getting ready to expand their operations to Hollywood.

The catastrophes that occurred. To begin, however, there is the process of trial and error, as well as the approval of items that are, at best, only partially developed. There have been images of Prime Video’s Korean dramas “My Boy is Cupid,” “True to Love,” and “The Beat of My Heart” that have been shared on social media for a few days now. These dramas have not only been released in Korean with subtitles, but they have also been dubbed into Spanish with the assistance of artificial intelligence. The Spanish dubbing comes with a neutral accent that is reminiscent of other eras of Latin dubbing. The effect is awful, as this or this Twitter account has pointed out: intonations that are unimpressive, sentences that contain errors, and, of course, a complete and utter absence of agreement between the visual and the sound.

Dubbers already on the offensive. Traditional voice actors and unions have, of course, begun to make ideas of all kinds. Some of these proposals have been prompted by the letter that Scarlett Johansson signed, in which she declared her opposition to OpenAI using her voice for an artificial intelligence assistant system. The actor Masumi Mutsuda discusses the demands that dubbing unions are currently putting in place regarding traceability and transparency. These demands include the requirement that it is possible to locate the origin of the voices that have been taken as a model and that the voices that are intended for a job are not intended for training voice models.

I have been around long enough to witness the continuous development of video games. In addition, they encountered a number of new significant benchmarks and criteria. The creation of voices by computers is not a novel concept. Other games, such as Berzerk, featured it. Prior to the implementation of sampled sounds, it had a rather robotic quality. During the time of laser discs, compact discs, and DVDs, many video games were able to include digitized versions of sound, voice, and music. When we go back to Lucasart games, we remember that they featured real voice acting rather than just text on screen, which made them feel more like cartoons. That, however, was not the end of it. There was even the possibility of digital people. Under a Killing Moon and Wing Commander III are two excellent examples of this in action. We suddenly had games that were similar to live-action movies.

In contrast to this, however, there was another technology that was rapidly developing, and that was graphics that were generated in three dimensions. In the beginning, it had a solid blocky appearance. However, over the course of many years, it has dramatically improved. Despite the fact that it is not yet completely lifelike, it is coming closer and closer. Take a look at the Final Fantasy series to see how it has developed over the years from a two-dimensional game to a game that is more like a movie.

Throughout all of these years, I have been wondering why they have not worked on developing a method of modeling sound and voices, similar to what they have done with 3D graphics presently. Perhaps they did, but the news did not make it to the mainstream media because it was so terrible. Twenty-nine years have passed, and all of a sudden, developers are experimenting with voice generated by artificial intelligence. And a great number of individuals have become enraged as a result of it. Voice actors are going on strike in an effort to save their jobs, and there is a lot of talk going on across all forms of media.

But wouldn’t you say that this is merely another notable achievement in the ever-evolving games? In Cyberpunk 2077, what is the difference between a 3D-generated version of Keanu Reeves and an artificial intelligence-generated version of his lines that are taken from the script? Much more dynamic responses are now possible as a result. For instance, there are a variety of circumstances that might have an effect on the voice, such as consuming food while speaking or being weary while the same lines of speech are being delivered. Without the need to store the audio for each individual circumstance. At this point in time, artificial intelligence is in its infancy, much like 3D was in the 1990s. There are triangles on top of triangles. It is possible that it will undergo a transition from simply blending previous work to actually being able to produce speech that is fully generated and includes parameter input. In my opinion, this is the first step right now. In the past, it was similar to how putting images and skins on blocks in early 3D technology was.

I am aware that the entire topic around artificial intelligence has an impact on a great number of people and areas of our planet. The possibility can be terrifying and transformative to one’s life. Video games, on the other hand, are all about technology. Why is it that the use of new technology to create new and exciting games should be considered controversial?

One of the most notable voice actors in the gaming industry has expressed their discontent with a substantial yet unclear agreement about the usage of artificial intelligence-generated voices. This new debate has erupted in the quickly evolving gaming field. A contentious discussion is taking place between actors and industry stakeholders around this technology, which has the potential to completely transform the way voice acting is performed in video games.

SAG-AFTRA, the United States actors’ organization, and Replica Studios, an artificial intelligence startup, have come to an agreement with the intention of ensuring that union members receive “fully informed consent and fair compensation.”

However, a significant number of voice actors, who are already concerned about the possibility of artificial intelligence taking over their jobs, have angrily opposed the accord. SAG-AFTRA reportedly sent an email to its members informing them that the arrangement was negotiated by a committee consisting of actors who have substantial experience in the gaming industry. This information was reported by the BBC. The union underlined that the contract was meant to cater exclusively to voice performers, giving terms for informed consent and suitable remuneration. There were also provisions for appropriate compensation.

This dispute may be traced back to the extended strike that SAG-AFTRA went on in 2023. During that strike, the labor union’s primary focus was on lobbying for protections against the use of artificial intelligence by film and television companies. grasp why the current agreement has prompted such a strong reaction among voice actors requires, first and foremost, a grasp of the historical backdrop.

This new arrangement is seen by a large number of voice actors as being in direct opposition to the objectives of that strike. It was mentioned by Sunil Malhotra, who is well-known for his work in the video games Fallout and Mortal Kombat, that he had joined in the strike the previous year in order to protect his career and not to make it easier for his AI-generated voice to be used.

SAG-AFTRA claimed in a blog post that it was ratified by members of the union’s voiceover community who are affected by this technology. The blog post was an announcement of the deal. Duncan Crabtree-Ireland, the lead negotiator for the union, emphasized the significance of safeguarding the rights of voice talent, particularly in light of the fact that gaming studios are increasingly investigating the uses of artificial intelligence for efficiency. He argued that the agreement ensures that members’ voices and acts are used based on their informed consent and that they are compensated fairly for their participation.

https://www.ictdemy.com/csharp/csharp-forum/alpha-cream-greece–6656de627c086#goto7342

https://www.saatchiart.com/art-collection/Alpha-Cream-Anaskopisi-kai-tmi-Prositi-anakoyfisi-apo-ton-pono-stis-arthroseis-Greece/2477337/807869/view

https://lavish-population.unicornplatform.page/alpha-cream-greece/

https://www.intelivisto.com/forum/posts/list/289935.page#432989

https://500px.com/photo/1093946699/alpha-cream-anaskophsh-kai-timh-prosith-anakoyfish-apo-ton-pono-stis-ar8rwseis-greece-by-alpha-cream

https://admisure.com/ask/297352/alpha-cream-%CE%B1%CE%BD%CE%B1%CF%83%CE%BA%CF%8C%CF%80%CE%B7%CF%83%CE%B7-%CE%BA%CE%B1%CE%B9-%CF%84%CE%B9%CE%BC%CE%AE-%CF%80%CF%81%CE%BF%CF%83%CE%B9%CF%84%CE%AE-%CE%B1%CE%BD%CE%B1%CE%BA%CE%BF%CF%8D%CF%86%CE%B9%CF%83%CE%B7-%CE%B1%CF%80%CF%8C-%CF%84%CE%BF%CE%BD-%CF%80%CF%8C%CE%BD%CE%BF-%CF%83%CF%84%CE%B9%CF%82-%CE%B1%CF%81%CE%B8%CF%81%CF%8E%CF%83%CE%B5%CE%B9%CF%82-greece

https://freead1.net/ad/5161648/alpha-cream-greece.html

https://hellobiz.in/alpha-cream——greece-997427020

https://publuu.com/flip-book/525884/1178261

https://archive.ph/vIZq8

https://archive.fo/vIZq8

https://indspire.microsoftcrmportals.com/en-US/forums/general-discussion/dc61cbae-a01d-ef11-989a-000d3a84ad0c