When i am trying to use batch generation, i got this
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.
I have a problem. I created the csv files and wanted to drag them to the Batch Synthesis window, but when I do that I get the message: Error: game_id "fallout4" does not match any available games. Why?
Thank you the tutorial is very helpful. Is there a patch for f4_batch_prep to handle languages other than english. I'm trying to make german dialogs and the plugin doesn't handle the german umlauts, it turns an ö into an ö for example.
Could this guide, and the plugins you mentioned and linked to, be used to add voices generated with ElevenLabs, instead of by xVASynth? Like, would there be a way to replace the .wav files that xVASynth generates with ones generated and downloaded from ElevenLabs before the lipsync generation process occurs, so the ElevenLabs .wav files could be lipsynced instead, and then combined and turned into .fuz files by this tool?
So.... I followed the guide. I'm pretty sure I have xVA Synth set up correctly. But when I convert my .csv files, the only output I get are .wav viles. No lip or fuz files. What am I doing wrong?
Update: Nevermind. I'm dumb and can't read instructions. Problem solved.
Hi radbeetle, I wondering if you can help me out? I'm stuck between steps 3 and 4. Seems like I didn't do the steps right to make a .csv file in order to put a .csv in batch synthesis.
I'm not sure what to title the files directory and the output file name. For audio format do I just use wav? I have 8 excel files when I run conversion which are located in This PC> Desktop > xVASynth_2.1> resources > app >plugins > f4_batch_prep.
The mod I'm trying to have a voiced default Nora player character is CSEP Presents All Americans.
I solved my issue by deleting the rows that have no text in the excel file. The guide worked and is easy to follow. Although there was some dialogue that still went unvoiced. I'm assuming the reason is either I missed something in the object window or these were one of the quests that supposedly have 0 dialogue lines.
Well I wish you the best in that your power outage issue is solved. I'll probably ask more questions or for more advise thru DMs or thru discord in VAsynth discord.
17 comments
When i am trying to use batch generation, i got this
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.
I created the csv files and wanted to drag them to the Batch Synthesis window, but when I do that I get the message: Error: game_id "fallout4" does not match any available games. Why?
Is there a patch for f4_batch_prep to handle languages other than english. I'm trying to make german dialogs and the plugin doesn't handle the german umlauts, it turns an ö into an ö for example.
Update: Nevermind. I'm dumb and can't read instructions. Problem solved.
I'm not sure what to title the files directory and the output file name. For audio format do I just use wav? I have 8 excel files when I run conversion which are located in This PC> Desktop > xVASynth_2.1> resources > app >plugins > f4_batch_prep.
The mod I'm trying to have a voiced default Nora player character is CSEP Presents All Americans.
I've made a note to come back and help you out with this but if you don't hear from me by the weekend just shoot me a DM.
You do want to use WAV yes, I can answer that off the top of my head at least!
Well I wish you the best in that your power outage issue is solved. I'll probably ask more questions or for more advise thru DMs or thru discord in VAsynth discord.
Thank you, we really got walloped over here, it sucks.
I'll be happy to help out as I can and I'm sure the folks in the discord will too!
This includes Nora.