|
Author
|
Message
|
|
4804419
|
4804419
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Years Ago
Posts: 11,
Visits: 33
|
Do you consider adding ai text-to-speech?I'm currently using plotagon for animation, I found their built-in scene templates, then add characters, and add ai text-to-speech to each character at the same time, for me, a person who can't do animation at all, it's really too simple.So, in the next version, do you consider adding ai text-to-speech? As a beginner, I think your editing track is too complicated. If you can consider a better experience, the effect will be better, such as a track layout similar to video editing software, which is divided into timeline, sound, props, etc. If you There is a better solution, that's better
|
|
|
|
|
Peter (RL)
|
Peter (RL)
Posted 3 Years Ago
|
|
Group: Administrators
Last Active: 2 Years Ago
Posts: 23.1K,
Visits: 36.6K
|
Adding AI text to speech isn't in the immediate plans, however Cartoon Animator already allows you to import audio files to create automatic lip-sync. This means you can use one of the available AI services like Replica to create the voice audio and just import that.
Peter Forum Administrator www.reallusion.com
|
|
|
|
|
animagic
|
animagic
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 4 Months Ago
Posts: 15.8K,
Visits: 31.4K
|
With specialized companies already making steady improvements in AI text-to-speech, there is little point in RL devoting any development time to that effort. And frankly, doing animation takes effort, if you want to make anything worth watching. If you want very simple there are some softwares aimed at creating animation for explainer videos.
|
|
|
|
|
4804419
|
4804419
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Years Ago
Posts: 11,
Visits: 33
|
There is a problem here. After I import an audio file, if there are multiple conversations in the entire audio file, how do we solve the problem of different characters' mouth shapes? And if we add an ai text-to-speech interface (of course, you can charge extra, you know that ai text-to-speech is very cheap now), when we do animation, if there are multiple characters in the whole scene, we only need to click on a certain character Is it more efficient to create animation by adding voice content?What's bothering me right now is: When I need multi-person dialogue when making animations, I generate all the dialogue script audio at one time, so how do multiple characters in the scene match mouth shapes + movements at this time? So this is also when I use plotagon, adding audio dialogue content directly to a single character solves my problem. So is there a good solution for similar problems in CTA5, how to add audio efficiently, suitable for dialogue with characters
|
|
|
|
|
AutoDidact
|
AutoDidact
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Months Ago
Posts: 2.1K,
Visits: 13.6K
|
With specialized companies already making steady improvements in AI text-to-speech, there is little point in RL devoting any development time to that effort. .Agreed!! AI voices have come so far in the past year that it is no longer even a concern for solo animators. I had sworn I would never create another feature length solo animated film after the torturous six years making “Galactus Rising” But AI voices ( other AI assets) changed my mind and I have 39 minutes completed on my Iclone exclusive “ghost in the shell” fan film that I started back on June 7th
|
|
|
|
|
4804419
|
4804419
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Years Ago
Posts: 11,
Visits: 33
|
In fact, what I want to suggest is: how to make animations more efficiently and add voices to animated characters, because when we do animation, there are multiple characters in the animation, and we need to dub each animated character, or between characters and characters Dialogue between, usually in this case, if we need to solve the dubbing problem, do we need to find the corresponding personnel to solve it? I think this is a difficult problem. If we use the text-to-speech function of ai, when we create multi-character dialogue dubbing, we only need to click on the corresponding character to add sound or add audio dialogue, so that we can create animation Will it be more efficient? The traditional method is that we first generate audio from the text, and then match the mouth shape. If there are multiple characters in a 10-minute animation, we need to generate another audio for each character. This efficiency It will be very slow. If we directly use the text-to-speech function of ai in the software, we only need to do the character dialogue scene when we are doing animation, and click to add the text-to-speech function of ai on the corresponding character, so that the efficiency will be more efficient. high
But the cta official said that this is not in the scope of consideration. I am currently researching blender, which has many plug-ins, and I hope to solve the problem of dubbing efficiency. Because my daily job is to do youtube animations, my animations will have dialogue scenes with multiple people, how to efficiently dub different people is a problem
|
|
|
|
|
4804419
|
4804419
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Years Ago
Posts: 11,
Visits: 33
|
Although your video uses AI dubbing, it looks very stiff and the characters have no movement. I have used a simple plotagon to make animations. They add the dubbing of the character's dialogue. Just click on the character to add text and it will automatically become audio. , great, efficient way of adding dialogue audio
|
|
|
|
|
AutoDidact
|
AutoDidact
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 2 Months Ago
Posts: 2.1K,
Visits: 13.6K
|
the characters have no movement. So the two male characters suddenly getting up from their seats in shock over the violent news feed is “no movement”?? and the female swaying slightly from side to side and actually looking at her holographic screen display during her briefing is “no movement”??
OK .. But I don’t believe in having my characters over gesturing wildy with flailing canned hand/head gestures during expositional dialogue as I see in those plotagon clips, it is a useless visual distraction particularly when you are using canned & repeated gestures as Plotagon clearly does. probably useful for cranking out a bunch of disposable youtube shorts though. I am currently researching blender, which has many plug-ins, and I hope to solve the problem of dubbing efficiency. Because my daily job is to do youtube animations, my animations will have dialogue scenes with multiple people, how to efficiently dub different people is a problem As a Blender user myself I can tell you with absolute certainty that your lipsync options both native and third party are utterly Abysmal or painfully complicated compared to Iclone ,CTA or even Daz studio. Notice how you see almost Zero Multi-character, dialogue heavy animated productions coming from the Blender user community. The reality is that when you have dialog heavy Character driven stories with multiple characters you will have to do ALOT of pre planning and will not always be able to have a linear work flow. Be prepared to make camera cuts and assemble the conversational clips in your video editor. I have CTA4 and MOHO 13.5 Both softwares give you the option of moving the voice dialogues to any point in the timeline as characters have conversations. however if you study actual filmmaking you will learn that it is rare to have a wide shot of multiple characters in frame all speaking in turn as many conversations you see in movies have camera cuts between the actors often shot on different days entirely.
But again if one needs quick disposable animated youtube vids Plotagon should do fine.
|
|
|
|
|
animagic
|
animagic
Posted 3 Years Ago
|
|
Group: Forum Members
Last Active: 4 Months Ago
Posts: 15.8K,
Visits: 31.4K
|
Note: I don't use CTA, but iClone, but the approach should be similar. I don't really understand the problem. What I do is create an audio file for each character's dialog. I do the timing in advance adding silence to moments where the character doesn't speak. I then apply the the timed dialogs to my characters and everything falls in the place. I then add body animations as needed. But, as AutoDidact points out, there is no need for exaggerated motions.
|
|
|
|