Hello, I'm working on an Arabic model for this project. I'm planning on adding Arabic characters support as well as having the software synthesize audio that speaks Arabic. So far I'm trying to do the following :
Modifying synthesizer/utils/symbols.py to include Arabic characters and numbers and I must have successfully done that.
Running encoder_train, vocoder_train and synthesizer_train with the corpus dataset to generate pt files for each which will server as the new pretrained model for Arabic language
The final step which consists of running the toolbox to check if the software functions well.
Hello, I'm working on an Arabic model for this project. I'm planning on adding Arabic characters support as well as having the software synthesize audio that speaks Arabic. So far I'm trying to do the following :
Downloading the corpus Arabic dataset linked here : http://en.arabicspeechcorpus.com/arabic-speech-corpus.zip
Modifying synthesizer/utils/symbols.py to include Arabic characters and numbers and I must have successfully done that.
Running encoder_train, vocoder_train and synthesizer_train with the corpus dataset to generate pt files for each which will server as the new pretrained model for Arabic language
The final step which consists of running the toolbox to check if the software functions well.
Is there anything else that needs to be done???