Can we just use FastSpeech for inference as baseline result

[Maoshuiyang]

Hi Keon, thanks so much for sharing this wonderful project. I am wondering can we just use the FastSpeech part for inference? Looking forward to your reply

[keonlee9420]

Hi @Maoshuiyang , thanks for your attention. Of course you can use it as you said, by modifying some part of code, but the better way is to check out this repo: https://github.com/keonlee9420/Comprehensive-Transformer-TTS. It contains the same model architecture of FastSpeech used in this repo and the model is only aimed to generate speech by itself, rather than used the output as an auxiliary input to another model. Hope it helps!