StyleTTS2
StyleTTS2 copied to clipboard
How do I use inference() method for Hindi language?
I am trying to use TTS for devanagari script but I'm getting an error
Traceback (most recent call last): File "C:\StyleTTS2\tts.py", line 7, in other_tts.inference("कैसे हो आप", output_wav_file="another_test.wav") File "C:\StyleTTS2\virt\lib\site-packages\styletts2\tts.py", line 273, in inference pred_aln_trg[i, c_frame:c_frame + int(pred_dur[i].data)] = 1 IndexError: invalid index of a 0-dim tensor. Use tensor.item() in Python or tensor.item<T>() in C++ to convert a 0-dim tensor to a number
I observed in other TTS like XTTS v2 there seems to be an in-built option that lets you select which languages' text we are giving it to process
Is there an ibuilt option? Do I have to use PLBERT for this in a special way i'm not aware about?