Open Okkband opened 6 months ago
Hi- 13 sentences are not sufficient for training the model. You need several hours of audio recording. You don't need to train 10000epochs. Just check the checkpoint's performance and if its good then you can stop in the middle
@Zengyi-Qin Are the pre-trained models English only? People have been training with Chinese data but not having good result.
@Zengyi-Qin Are the pre-trained models English only? People have been training with Chinese data but not having good result. I have to trained the model with 1hour's sentences for four days. but I cannot get a good results.
Please add some instructions how to prepare data and train to another languages (I would like to train Czech language), it will helpful for begginers. Thank you.