[Released] Anilip 2 [Commercial]
Dobit
Posts: 197
To celebrate Anilip's 1 year anniversary, we will be offering a brand new upgrade in the next few days.
So many new things have been added that we are upgrading to version 2.
There was a long todo list we worked on last year.
Here are just the most important innovations:
- Anilip 2 now supports many more voices.
- "Speech Recognition" of voice recordings
- "Text To Speech"
- Cloud services (Transcribe and Polly from Amazon)
- Generic JSON interface.
- A new viseme model allows a wide range of speech styles (from whispers to screams). Lip animators can now also implement emotional speech.
- Our new tabular lip animation tool makes it easy to translate even lengthly animations.
- The configuration of the lip animation was exposed for all figures in an XML file. Now also figures that are not offered by us can be adapted.
- We put a lot of work into documentation. 4 tutorial videos, 1 manual, inline help.
A first impression is here.
popup_2.jpg
1000 x 1291 - 211K
popup_3.jpg
1000 x 1300 - 220K
popup_1.jpg
1000 x 1300 - 176K
popup_4.jpg
1000 x 1300 - 286K
popup_5.jpg
1000 x 1300 - 203K
popup_6.jpg
1000 x 1300 - 193K
popup_7.jpg
1000 x 1300 - 280K
Post edited by Dobit on
Comments
This looks to be useful.
Oh nice, I will be wishlisting this, I was finally able to get the first one recently and haven’t had time to try it out or find the program for that matter but the new features sound awesome :)
Is this a windows only app or can it be used on the Mac platform?
This is a Windows only app.
Why NOT for MAC OS ?
Many Mac user like me will buy that imediately !
At the moment we lack on MAC developers. But good to know the demand.
The integration of AWS Polly sounds interesting, but I can't find any clue about the generated audio files... Can the audio created with Polly in the plugin be used in commercial products?
I am not a lawyer and you should read the AWS legal documents. But their use cases in the polly faqs mention commercial projects.
Yes, it is what I found too, pretty confusing to me. I asked you because I thought you could know better since you used their API.
Thanks for the answer!
I bought this plugin and I'm having serious problems with it, I'm using the supplied test scene with no edits, and following the turorial video exactly, but the viseme animation only ever produces one single viseme "ç" regardless of what or how much text I enter.
You try "Text to Speech" or "Speech Recognition"? Please check the log file for error messages. (Help/Troubleshooting/View Log File...). Send a screenshot of your "Settings/Text To Speech Settings" Tab (you can use the dobit support mail).
We use Polly sound files commercially. The AWS Service Terms have not many restrictions for audio files. Here I found a short discussion about Polly files and copyright.
Is there a way to smooth transitions, that is, to create intermediate incremental keyframes between frames?
Yes, that's how it works. The transition between the individual phases of a viseme (keyframes start, apex, end) is generated automatically and smoothly. If you want to configure it, you can change the parameters of the TCB splines. As part of the delivery, we have a video explaining the Viseme model. There will also be discussed on the generation of key frames. More smoothing can be achieved by increasing the visemes ascent and or descent time and shorten the apex time. (I hope I understood your question ).
OK, I'm getting this.
Log file:
2019-01-09 21:19:02.015 WARNING: libpng warning: iCCP: known incorrect sRGB profile
2019-01-09 21:21:07.246 DEBUG: Anilip 2: Phoneme with ID '4' unknown
2019-01-09 21:21:07.246 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.359 DEBUG: Anilip 2: Phoneme with ID '104' unknown
2019-01-09 21:21:07.359 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.509 DEBUG: Anilip 2: Phoneme with ID '601' unknown
2019-01-09 21:21:07.509 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.569 DEBUG: Anilip 2: Phoneme with ID '108' unknown
2019-01-09 21:21:07.569 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.659 DEBUG: Anilip 2: Phoneme with ID '111' unknown
2019-01-09 21:21:07.659 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.719 DEBUG: Anilip 2: Phoneme with ID '865' unknown
2019-01-09 21:21:07.729 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.809 DEBUG: Anilip 2: Phoneme with ID '650' unknown
2019-01-09 21:21:07.809 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:07.869 DEBUG: Anilip 2: Phoneme with ID '4' unknown
2019-01-09 21:21:07.869 DEBUG: Anilip 2: No viseme class found for phoneme ''
2019-01-09 21:21:23.987 WARNING: cloud\dzcloudtasknotifier.cpp(178): peer performed orderly shutdown errno=0
2019-01-09 21:24:23.166 WARNING: cloud\dzcloudtasknotifier.cpp(178): peer performed orderly shutdown errno=0
2019-01-09 21:27:40.820 WARNING: cloud\dzcloudtasknotifier.cpp(178): peer performed orderly shutdown errno=0
I've attached settings.
There is a problem with the UK Locale. Anilip 2 uses the wrong phoneme set for this. We are currently working on a solution and will provide an update in the next few days.
Thanks for the quick response. So if I switch to US English, should be different? I will try. Any other changes? Or is this more at a system level on the PC?
An intermediate solution is to switch to a US English voice. The manual gives some advice how to add a new language in the windows settings. There are phoneme/viseme differences between US and UK. I attached an example for "Hello World".
I am also having this issue attempting to use UK voices. US voices are working for TTS. Hoping for an update soon, thanks.
OK, will try. It may or may not be relevant that Daz crashed last night at some point after trying to use Anilip, so I'll watch out for that as well. I thought Windows had updated but I don't think that was what happened.
On further investigation a reboot.
We have made new Anilip 2 binaries to fix the UK Locale issue. This is an unofficial version, which we send to users upon request. Please send an e-mail to Dobit support and we will send you information for download. We will also release a new version via Daz, but this will take some time, because we collect user feedback first.
Thank you! UK voices now appear to be working with the fix.
Well, heck... I think this is what I've been looking for. I didn't even know Anilip 1 was a thing. It's gone now, in favor of 2 right?
This doesn't animate anything but the mouth, right? You do the rest of the face, the neck and hear, eyes and so on, to sync right? And are there controls to exaggerate the phonemes more or less? For instance, in the Crispin's Day speech, there are parts where much more exaggerated mouth movements would (in my opinion) work better...
I think what I'll do is use text to speech to do the animation, then sync voice to that, as the voice actors watch the animation - that sound right?
Yes, Anilip 2 Is the replacement of Anilip.
Anilip 2 specializes in visemes. It animates the mouth and the jaw. In the promo video we used Animate 2 to sync with all the other animation parts you menttioned.
The biggest difference to the previous version is the viseme model. This allows selected visemes to be emphasised or weakened (Even extremes like screaming or whispering are possible)
The voice of Henry V was first recorded by a professional speaker and then animated. We used Amazon Transcribe for speech recognition. But your workflow sounds Ok as well.
Dobit, is Anilip 2, by any chance, going to be on sale this month (50% off or something like that)? I'd like to get this to use for a Daz-animated music video I'm making for a musician friend of mine on my Shadow Cloud PC (which I'm subscribed to until the 28th), but the current price of Anilip 2 is too far above what I can afford.
Unfortunately, we can not predict that because Daz Marketing plans these discount promotions. But a discount of 50% is unlikely.
Nuts... Well, thank you for your answer. Just out of curiosity, do you have any advice on how to get the best results out of DAZ Studio's built-in lip sync plugin? I don't know if you have experience with the lip sync plugin, but since you made Anilip 2, I'm assuming you've used it in the past. I have had mixed results with it (some parts look believable, some don't... in the same animation sequence no less). Thank you again.
Daz Studio does not have a built-in LipSync module.You say that you want to make a music video. If your character is to sing to music, it may be possible to animate the viseme "by hand". That works, if not too much text is to be set. The "Parameter" pane lists all visemes of your Genesis 8 or Genesis 3 figure in the path "Pose Controls / Head / Visemes". If you spread the viseme over the timeline, you can implement a simple lip animation.
Thank you for the advice, but I found a (free) solution using papayago lip sync software and a daz studio script.