Christian Schäfer
Christian Schäfer
Locally I run phonemizer=2.2 without any issues.
Thx for looking into it. I just merged a fix that should solve the problem (seems that phonemizer >=3.0 changed the interface into expecting a List[str] instead of Union[List, st]).
Hi, the examples are vocoded with the V1 version of HiFi-GAN which exhibits a lot less artefacts than the V3 version. That being said, the current uploaded model is a...
Hi, looks like the train set is empty, probably because with 21 files everything lands in the validation set (standard config for n_val=200). 21 files is surely not enough to...
Hi, glad you like it. How much data do you have? 1. If you synth across multiple sentences then you would have to produce training data with multiple sentences as...
Nice. Sounds quite good already, but imo the WaveRNN could still improve a bit (the gnarling/hissing) - how many steps is this for vocoder and tts? The hissing could also...
Hi, very cool. This is something on my list, I will also try to train multispeaker models which I hope will improve the pitch prediction. I am pretty sure that...
Hi, thanks for pointing it out. I personally don't use espeak anymore, but I will look into it - there could be problems with the multiprocessing imo.
Seems like the fix is easy, here you go: https://github.com/as-ideas/ForwardTacotron/pull/84
Hi, what specifically do you want to do with SSML? Including pauses for example is pretty straightforward, as would be pitch modifications. However, you would have to implement it yourself...