How i can fix this problem? It gives me this message when i trying to use batch generation
Traceback (most recent call last): File "server.py", line 335, in do_POST File "python\xvapitch\model.py", line 326, in infer_batch File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 223, in infer_advanced return self.infer_using_vals(logger, plugin_manager, cleaned_text, text, lang_embs, speaker_embs, pace, None, None, None, None, None, None, pitch_amp=pitch_amp) File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 292, in infer_using_vals x, x_emb, x_mask = self.text_encoder(input_symbols, x_lengths, lang_emb=None, stats=False, lang_emb_full=lang_emb_full) File "torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 602, in forward x = torch.cat((x_emb, lang_emb_full), dim=-1) RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 27 but got size 1 for tensor number 1 in the list.
Sorry for the dumb questions but, does this still work on v3? Also, it says "automatically" but I assume that just means this handles the actual generation rather than me having to build them myself from scratch but I still have to run something to generate the files, how do I do that? It's been a long time since I added my voice files and I've forgotten a few things.
Can someone help me i have updated to version 3.0 now i am getting this error on batch conversion
Traceback (most recent call last): File "server.py", line 335, in do_POST File "python\xvapitch\model.py", line 326, in infer_batch File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 223, in infer_advanced return self.infer_using_vals(logger, plugin_manager, cleaned_text, text, lang_embs, speaker_embs, pace, None, None, None, None, None, None, pitch_amp=pitch_amp) File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 292, in infer_using_vals x, x_emb, x_mask = self.text_encoder(input_symbols, x_lengths, lang_emb=None, stats=False, lang_emb_full=lang_emb_full) File "torch\nn\modules\module.py", line 1194, in _call_impl return forward_call(*input, **kwargs) File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 602, in forward x = torch.cat((x_emb, lang_emb_full), dim=-1) RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 25 but got size 1 for tensor number 1 in the list.
- I update xVASynth to 2.2.1 (2.1.0 + the small 2 2 1 patch) cos, yeah, your plugin cant be ticked on without my earlier version. - Updated a bunch of voices models versions and new ones. - Reinstalled and enabled your plugin over everything again to get sure. - Closed and relaunched the app a few times to confirm settings...
- wav, and json files are sucessfuly generated as i m used to (i left the json because it was the default configuration. the app mentions "needed for editing". If this editing is the letter and silence manipulation, thats what i do sometimes so its fine...)
Just no lip file.
Audio option : 44100 Hz + pcm/16 bit (that was default) About the plugin settings, i changed the "Game" settings to Skyrim/Fallout 3 etc since it was Fallout 4 by default. Delete lip files, delete wav, Use multiprocessing and skip existing files are unticked Make fuz files is now ticked (as a test, it wasnt on a few first tries i just wanted the lip files added) Num process is 10 (default i guess.)
I cant get why it doesnt work.
The only "unusual" thing is that i made my own output path, closer to the root folder of Xvasynth folder (in it actually) so i can find my files more easily and also make sub folders to sort my work without worrying about too long path the system cant deal with ^^ Wave and json always worked there, and still work inside the auto generated folders like "sk_[voicemodelname]"
Could you help ? Your tool is my only hope to avoid the only way i was able to generate lip files so far = a super heavy (and long) process with the CK = treat ALL the game orphan files (wav with correct format without an existing lip file). the CK would browse every single ESP \ DIAL records *sighs*. Thats super long when you want to include 4 - 5 audios :/
If you mean do i have FaceFXwrapper installed exactly how and where it should in your description text, yes i do ! Its the latest, 0.4 version. I had it already from my SSE CK fix version install, in another new place now so (in your lip_fuz folder under ...\plugins). as you requires it. It seems correctly installed for XvaSynth and your plugin since the latest version of it mentionned a Fallout 4 option added. And i have two choices in XvaSynth settings menu, as i described above, about what game format i want for the outpuit (Fallout 4 or Skyrim + other games).
Could it be a security issue about this exe (admin / compatiblity mode or something else ?). Cant remember any warning or authorization request for this exe ever tho. Its a bit old but i cant remember making some install description of this exe work for CK SSE fixed version neither. my patience on the CK was, and is still super low so i think i gave up at some point. I m not really sure... As i said above, i can still generate working lip files from the mass / all game process ( CK > Gameplay > Facial Animations)
I've had rare occasions where the facefxwrapper file didn't actually work on some people's PCs. You could try just running that exe (without xvasynth) from the commandline, to test if it actually works on your PC
Did you find a solution? Having same issue. lip and fuz files not generating. wav and json files are. ran fxwrapper from command line. program ran but looking for very specific parameters.
Hithere.. Just reading your question now @joemarotta, sorry. Well no, not yet. since i didnt tried anything about that recently.
@DanRuta & murrayj : Since the CK generate working lip/fuz files with the global (whole game ) check and process, doesnt it means that FaceFXwrapper doest work with certain conditions ? (works for me as i said earlier, just not with XVASynth) Or is it another process ?
Exact same issue on V2.3. All t's dotted and i's crossed but the Status just freezes at Running as soon as I start Batch Synthesis. Nothing generates at all. Disabling the plugin, I at least get the output wav data that I can use, but the lip/fuz is just not in the plan it seems.
I noticed an 'ERROR.txt' file created in the main app folder, and it contains this:
Traceback (most recent call last): File "run_lipfuz_gen.py", line 80, in <module> ValueError: invalid literal for int() with base 10: 'Skyrim'...so, i would say there is something clearly wrong.
I've got the exact same problem. Followed the installation guide exactly. I was able to enable the App just fine. When I generate a voice file, I get the voice and the json. That's it. I'm using VA2.3.0.
Give Feedback
53 comments
Any suggestions?
EDIT: I found a solution by adding both FaceWrapper and FonixData file to the plugins folder in Xsynth.
Traceback (most recent call last):
File "server.py", line 335, in do_POST
File "python\xvapitch\model.py", line 326, in infer_batch
File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 223, in infer_advanced
return self.infer_using_vals(logger, plugin_manager, cleaned_text, text, lang_embs, speaker_embs, pace, None, None, None, None, None, None, pitch_amp=pitch_amp)
File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 292, in infer_using_vals
x, x_emb, x_mask = self.text_encoder(input_symbols, x_lengths, lang_emb=None, stats=False, lang_emb_full=lang_emb_full)
File "torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "D:\Modding\XVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 602, in forward
x = torch.cat((x_emb, lang_emb_full), dim=-1)
RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 27 but got size 1 for tensor number 1 in the list.
Also, some tutorial steb-by-step on how to use it would be welcome :)
Traceback (most recent call last):
File "server.py", line 335, in do_POST
File "python\xvapitch\model.py", line 326, in infer_batch
File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 223, in infer_advanced
return self.infer_using_vals(logger, plugin_manager, cleaned_text, text, lang_embs, speaker_embs, pace, None, None, None, None, None, None, pitch_amp=pitch_amp)
File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 292, in infer_using_vals
x, x_emb, x_mask = self.text_encoder(input_symbols, x_lengths, lang_emb=None, stats=False, lang_emb_full=lang_emb_full)
File "torch\nn\modules\module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Steam\steamapps\common\xVASynth\.\resources\app\python\xvapitch\xvapitch_model.py", line 602, in forward
x = torch.cat((x_emb, lang_emb_full), dim=-1)
RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 25 but got size 1 for tensor number 1 in the list.
anyone have any idea's what this all means
Thanks
- I update xVASynth to 2.2.1 (2.1.0 + the small 2 2 1 patch) cos, yeah, your plugin cant be ticked on without my earlier version.
- Updated a bunch of voices models versions and new ones.
- Reinstalled and enabled your plugin over everything again to get sure.
- Closed and relaunched the app a few times to confirm settings...
- wav, and json files are sucessfuly generated as i m used to (i left the json because it was the default configuration. the app mentions "needed for editing". If this editing is the letter and silence manipulation, thats what i do sometimes so its fine...)
Just no lip file.
Audio option : 44100 Hz + pcm/16 bit (that was default)
About the plugin settings, i changed the "Game" settings to Skyrim/Fallout 3 etc since it was Fallout 4 by default.
Delete lip files, delete wav, Use multiprocessing and skip existing files are unticked
Make fuz files is now ticked (as a test, it wasnt on a few first tries i just wanted the lip files added)
Num process is 10 (default i guess.)
I cant get why it doesnt work.
The only "unusual" thing is that i made my own output path, closer to the root folder of Xvasynth folder (in it actually) so i can find my files more easily and also make sub folders to sort my work without worrying about too long path the system cant deal with ^^
Wave and json always worked there, and still work inside the auto generated folders like "sk_[voicemodelname]"
Could you help ?
Your tool is my only hope to avoid the only way i was able to generate lip files so far = a super heavy (and long) process with the CK = treat ALL the game orphan files (wav with correct format without an existing lip file). the CK would browse every single ESP \ DIAL records *sighs*.
Thats super long when you want to include 4 - 5 audios :/
Thanks.
Its the latest, 0.4 version. I had it already from my SSE CK fix version install, in another new place now so (in your lip_fuz folder under ...\plugins). as you requires it.
It seems correctly installed for XvaSynth and your plugin since the latest version of it mentionned a Fallout 4 option added. And i have two choices in XvaSynth settings menu, as i described above, about what game format i want for the outpuit (Fallout 4 or Skyrim + other games).
Could it be a security issue about this exe (admin / compatiblity mode or something else ?). Cant remember any warning or authorization request for this exe ever tho.
Its a bit old but i cant remember making some install description of this exe work for CK SSE fixed version neither. my patience on the CK was, and is still super low so i think i gave up at some point.
I m not really sure... As i said above, i can still generate working lip files from the mass / all game process ( CK > Gameplay > Facial Animations)
Well no, not yet. since i didnt tried anything about that recently.
@DanRuta & murrayj : Since the CK generate working lip/fuz files with the global (whole game ) check and process, doesnt it means that FaceFXwrapper doest work with certain conditions ? (works for me as i said earlier, just not with XVASynth) Or is it another process ?
All t's dotted and i's crossed
Traceback (most recent call last):...so, i would say there is something clearly wrong.File "run_lipfuz_gen.py", line 80, in <module>
ValueError: invalid literal for int() with base 10: 'Skyrim'
Any ideas?